The Trouble With Book Groups

Before the parties, before the manifestos, before anyone was arrested for it, most modern revolutionary movements began the same way: a handful of people meeting to read something together. Russians called the form the kruzhok — literally "little circle" — and the nineteenth century produced them by the hundreds: clerks and students and minor officials gathering in someone's rooms to work through Fourier, Hegel, Feuerbach, later Marx, a chapter at a time, arguing about what they meant. The Petrashevsky Circle, the one that got a young Dostoevsky a mock execution and a Siberian sentence for attending, was in form indistinguishable from a book club. So were dozens of others that left no trace at all, because they talked for a season and then stopped meeting. The interesting question isn't why revolutionary reading circles failed to become revolutions. It's why the circle itself — the recurring meeting, the shared text, the argument that was supposed to sharpen into action — kept dissolving before it got anywhere, and what it took to build something that didn't.

The commitment is bigger than it looks

The kruzhok asked more of its members than it appeared to. The stated commitment was: read the text, come argue about it. The actual commitment was to sustain, alone and unwitnessed, the labor of working through difficult material — Hegel's Phenomenology, volumes of Marx — on a schedule set by other people's expectations rather than your own interest, while a job, a family, and the ordinary business of staying unarrested proceeded on no such schedule at all. Herzen described circle after circle in his memoirs animated by real intellectual hunger that simply could not survive contact with a life that had other demands on it. The people who stayed were disproportionately those with the least to lose by way of career or family — students, minor gentry with time on their hands — which meant the circles selected, from the start, for a membership unrepresentative of the society they hoped to change.

This is the same asymmetry any book group runs into, just with higher stakes. A weekly dinner or a chess club front-loads its obligation into the event; showing up is most of what's asked. A reading circle front-loads the obligation into the solitary hours beforehand, hours nobody else can see you skip. Add secrecy and the threat of arrest — reasons not to keep incriminating texts lying around, reasons not to write anything down — and the private, unenforceable nature of the commitment only deepens. Nobody could take attendance on the reading itself.

Reading well and arguing well are not the same skill, and neither is organizing

Circle discussions ran on the same hidden scoreboard any book club runs on, except the stakes made it worse: everyone in the room knew who had actually worked through the week's chapter of Marx and who was reconstructing an opinion from the discussion so far. Memoirists of the period — Kropotkin among them — describe circles dominated by one or two members with a real command of the material, doctrinal disputes that consumed entire evenings without anyone being persuaded of anything, and a chronic gap between people who could read theory well and people who could actually organize, print, distribute, or recruit. A circle could contain brilliant exegesis of Capital and produce nothing beyond the room it happened in, because interpreting a text and building an organization capable of acting on it are different competencies, and the kruzhok format assumed both would show up in the same people.

Lenin made exactly this diagnosis the center of What Is To Be Done? in 1902, and gave the disease its own name: kruzhkovshchina, "circle-ism" — the tendency of Russian Social Democracy to remain a scattering of local reading and discussion groups, each rediscovering the same debates from scratch, none accountable to any structure beyond itself, dissolved by arrests and reconstituted from nothing every time. His diagnosis wasn't that the circles read the wrong things or argued badly. It was that a movement made of independent circles, however sincere, could not accumulate anything — no continuity of leadership, no division of labor, no capacity to act in concert across a city let alone an empire. What he proposed to replace it with was blunt: a corps of "professional revolutionaries," a newspaper as an organizing skeleton binding scattered locals into one structure, and explicit hierarchy in place of the circle's easy, undirected equality. It is, read at this distance, a management diagnosis before it is an ideological one — the answer to why an association of well-read, well-intentioned people kept failing to become an organization capable of anything.

Bakunin was working the same problem from the opposite direction at almost the same moment, and his answer is worth setting beside Lenin's precisely because it starts from the opposite premise and still arrives at hierarchy. Bakunin had no more patience than Lenin did for a movement that stayed a scatter of local circles, each debating from first principles, none capable of acting in concert — but he could not, as an anarchist, propose a party or a state-in-waiting to fix it. What he built instead were secret brotherhoods: the Alliance of Socialist Democracy, and beneath it an inner ring he called, with unembarrassed candor, the "Invisible Dictatorship" — a small, disciplined, self-selected cadre meant to coordinate and steer the broader movement from inside it, without ever being visible as a governing body, since a visible one would have been a state by another name and therefore a betrayal of the whole project. It is a strange and telling piece of engineering: an anti-authoritarian who concluded that avoiding the circle's drift into nothing required exactly the mechanism the philosophy forbade — a leadership stratum, tighter and more secretive than anything Lenin proposed openly, just built to disclaim its own existence rather than publish itself as a newspaper and a central committee. Marx and the General Council expelled Bakunin from the International in 1872 partly on the accusation that the Alliance was doing precisely this, running a conspiracy inside a federation that was supposed to have no center at all.

Bakunin's brotherhoods solved the circle's drift problem the same way Lenin's party did — continuity, division of labor, a core that didn't dissolve when any one member was arrested — and paid for it the same way too, in the coin anarchists are supposed to refuse: an inner group deciding, unaccountably, what an outer group would be permitted to know or steer toward. The difference from Lenin's solution was smaller than either man's rhetoric admitted. Lenin built his hierarchy in the open and defended it as necessary; Bakunin built his in secret and denied it changed anything, which made it harder to correct when it inevitably produced the same complaints of manipulation and unaccountable authority that both men had set out, from opposite ends, to escape. The lesson isn't that anarchism failed where Bolshevism succeeded — the Bolsheviks' own hierarchy produced its own catastrophes soon enough. It's that the circle's underlying problem — no continuity, no accountable structure, no way to convert argument into coordinated action — was real enough that even a philosophy built on refusing hierarchy kept reinventing a version of it to survive contact with that problem, and the honesty of admitting the trade rather than hiding it is close to the whole difference between the two experiments.

The selection problem has no clean solution

A circle also had to decide what to read, and every method carried its own cost. Defer to whoever held the most doctrinal authority — the returned émigré, the one who'd actually gotten hold of a smuggled foreign edition — and the group became an audience for that person's positions, fine until a faction formed around a rival reading and the circle split, which Russian and later émigré Marxist circles did constantly and often permanently, along lines that mattered enormously to the people inside them and look, from outside, like the same recurring argument about who owned the correct interpretation of a text everyone claimed to have finished. Let the group rotate texts by consensus and the reading list drifted toward whatever was fashionable that season rather than whatever the movement's actual problems required, since consensus optimizes for offending nobody rather than for being right. Neither method reliably converged an argument into a plan of action, and plenty of circles never tried to converge at all — the reading and arguing were the whole point, self-improvement and camaraderie under cover of revolutionary purpose, and there is no shame in that except that it explains why so many of them left no further trace.

The IRA solved a different problem: infiltration

Lenin and Bakunin were both answering the same question — how does a scatter of circles accumulate into something continuous — and both answered it by concentrating authority somewhere, openly or not. The IRA's reorganization gives a third answer, because the problem it was built to survive wasn't drift, it was informers. A discussion circle can be infiltrated and merely embarrassed; an armed organization that gets infiltrated loses its members to a prison cell or a firing squad, so the design constraint that mattered most wasn't keeping the group coherent, it was making sure that no single capture could unravel the rest of it.

Michael Collins's reorganization after 1917, and the Provisional IRA's later "cell structure" from the early 1970s onward, solved this by doing almost the opposite of what Lenin and Bakunin did: instead of building a core that knew more and coordinated the whole, they built a structure in which no one below the top knew more than they strictly needed to. Volunteers were grouped into small active service units — four to eight people typically — nested under a company, battalion, and brigade hierarchy that existed on paper but was deliberately opaque sideways: a member of one unit usually couldn't identify the members of the next one over, and communication between cells ran through a small number of designated links rather than an open network anyone could map. Where a kruzhok's failure mode was too little structure to survive an arrest, the IRA's cell was engineered around the assumption that arrests were inevitable and had to be made survivable — compartmentalized so that breaking one member, even under interrogation, could only expose a handful of others rather than the whole organization.

The cost of that design was exactly the cost you'd predict from the book-group problem in reverse. A structure built to limit what any member knows also limits what any member can contribute to — the wide-ranging argument that made the old reading circles worth attending, the sense of a shared project being reasoned out together, has no place in a cell that runs on need-to-know. Members lower in the structure often had strikingly little visibility into strategy or leadership decisions, which bred exactly the kind of resentment and factionalism — visible at several points in the IRA's history, including its splits — that secrecy was supposed to prevent by other means, since a membership kept in the dark about the reasoning behind decisions has no way to distinguish sound strategy from mismanagement and tends, eventually, to assume the worst. Security bought at the price of the collective reasoning was the whole appeal of the circle to begin with.

A counter-revolutionary version of the same problem

Everything above assumes the circle is trying to move the country toward something it doesn't yet have. There's a mirror-image version of the problem, run by people convinced the country already had the right arrangement and has been talked out of it — restoration rather than invention, correction rather than overthrow — and it's worth asking whether the same organizational lessons hold when the direction of travel reverses.

The clearest American case is the John Birch Society, founded by Robert Welch in 1958, and its founding is unusually well documented because Welch was explicit about the model he was copying: the Communist Party's own cell discipline, which he'd studied as the thing to beat rather than the thing to fear indiscriminately. Birch chapters ran small — a dozen to twenty members, meeting regularly, working through a fixed course of material the way a kruzhok worked through Marx, except the assigned reading argued that the republic had drifted from its constitutional footing under communist influence and needed to be walked back rather than forward. The chapter structure gave each cell real cohesion and a clear, bounded task — write letters, staff a local campaign, run a "front" organization on a specific issue — while Welch's own council at the top kept the same kind of unaccountable, self-selected core that Bakunin's Alliance had, minus the excuse that revealing it would betray the philosophy. It was simply run like a business, because Welch had run one, and he saw no contradiction between the language of restoring liberty and the discipline of a hierarchy built to survive its own founder's departure.

The trade this made was the same one every organization in this essay has made, just wearing restoration's clothing instead of revolution's. A circle convinced it is defending something rather than building something recruits differently — older, more embedded in existing institutions, with more to protect and therefore more patience for paying dues and showing up to the next meeting than a student kruzhok's transient membership ever had — and that changed the group's most persistent failure mode from dissolution to drift into paranoia, a chapter that never has to prove its diagnosis against events because the whole point of restoration rhetoric is that the loss it's organized against is diffuse, ongoing, and never fully undone. Where a revolutionary circle's discipline problem is proving the future it wants is achievable, a restorationist circle's discipline problem is proving the past it's defending was ever as settled as the pitch requires, and Birch chapters that couldn't answer that question with anything but more meetings are exactly why the society calcified into a byword for overreach rather than becoming the mass movement its own cell structure was built to scale into.

That's the case this site's own name sits closest to, and it's worth saying so rather than hiding behind the historical examples. If social liberty is something America once had more of and is meant to find its way back to, rather than something to invent from nothing, then the organizing problem facing anyone who takes that seriously is the restorationist's problem, not the kruzhok's or the Bolshevik's: not secrecy, not surviving infiltration, but resisting the slide into a closed diagnosis that never has to be checked against whether the country is actually moving. The corrective isn't Lenin's newspaper or Bakunin's invisible dictatorship — there's no hostile state here forcing that choice — it's closer to what the Birch Society never managed: staying answerable to evidence that the direction is right, not just fluent in the reasons it must be.

Committees of correspondence, and the argument Jefferson and Adams never finished

America's own revolutionary circles predate the Russian kruzhki by a century, and they're worth setting alongside them because they're the rare case of the format actually working at scale. The committees of correspondence that Samuel Adams organized starting in Boston in 1772 were, in form, exactly the same thing as a kruzhok: small local groups, meeting to work out a shared position and put it in writing. What kept them from dissolving the way most Russian circles did was the thing in the name — correspondence. Each committee wrote to the others, regularly, and the letters did the job Lenin's newspaper would later do for the Bolsheviks: they turned a scatter of towns with no formal authority over each other into something that could act in rough concert, adopt shared language, and coordinate a colony-wide response, all without anyone at the center ever holding the kind of command Lenin or Bakunin insisted a movement needed. It's a genuine counterexample to this essay's running argument that circles need a concentrated core to survive — or, read more carefully, a case where the correspondence network itself was the core, distributed rather than concentrated, and durable enough that John Adams could later call the committees "the first Congress" in embryo, the structure that made the actual one possible.

What the committees didn't settle, and what Adams and Jefferson spent the last fourteen years of their lives arguing about in the letters they resumed after 1812, was what the thing they'd built was actually for going forward — whether it was a single act of founding that now needed protecting, or a standing license to keep refounding. Jefferson's side of that correspondence keeps returning to a version of the same claim he'd made to Madison in 1789: that the earth belongs to the living, that no generation has the right to bind the next with laws or constitutions it never consented to, that some measure of recurring upheaval — new conventions, rewritten constitutions, "a little rebellion now and then" — was not a failure of the republic but the maintenance it required to stay a republic at all. Adams, writing back from Quincy, wanted none of it. He'd watched the French Revolution take the same premise and turn it into the Terror, and his letters press Jefferson again and again on where, exactly, the recurring refounding stops — what protects the settlement they'd already built, with its checks and balances and its "natural aristocracy" of talent and virtue Adams thought every society produced whether it admitted to it or not, from being torn up by the next generation's confidence that it, too, had the right to start over.

That argument is this essay's revolutionary/restorationist split, run by the two men best positioned in American history to have it out properly, because both had helped write the founding and neither could be accused of not understanding what was actually at stake. Jefferson's position is the kruzhok's own logic pushed all the way to a governing principle: circles exist to keep the argument alive, generation over generation, because settled institutions rot exactly the way settled reading circles do, into a form nobody thinks to question. Adams's position is the Birch Society's better ancestor, minus the paranoia: institutions, once rightly built, are the hard-won thing worth defending against the next circle of confident young men convinced they can improve on it from first principles, and a republic that can't distinguish genuine correction from a plausible-sounding rebellion won't survive its own founding generation. Neither man won the argument, which is presumably why the letters are still read. A country's committees of correspondence solved the organizing problem; they never had to solve, and probably couldn't have, what the organizing was in service of once independence was won — and that's the question this site's own premise, that social liberty is a matter of America finding its right way, inherits from both of them rather than from either one alone.

What actually built something that lasted

The organizations that outgrew the circle stage — the Bolsheviks most visibly, but also the German Social Democratic Party, which ran its own network of workers' reading societies but bolted onto them a mass press, a trade-union apparatus, and electoral machinery that gave the theory somewhere to go — shared traits the founding circles never had. They accepted uneven commitment as a condition to design around rather than a moral failing to shame people out of: dues, defined roles, and a professional cadre who treated organizational maintenance as their actual work, not an unglamorous chore that fell to whoever happened to care most that month. They stopped expecting every member to be equally capable of both deep reading and effective action, and built division of labor instead — theorists, agitators, organizers, financiers, each doing the part they were suited to rather than everyone attempting the whole role. And they built continuity into the structure itself — a newspaper, a central committee, records and lines of succession — so the group could survive the arrest of any one member instead of dissolving with the room it met in, which is precisely what the loose, admirable, and largely evanescent kruzhki never managed.

None of that is what anyone imagines when they start a reading circle. They imagine the argument, the shared discovery, the text cracked open together — not the dues ledger, the org chart, the person willing to do the unglamorous work of keeping the thing alive between meetings. But the argument was never the hard part; put a room of people who've read the same difficult book together and they will find plenty to say, sometimes for an entire winter. The hard part, then as now, is everything upstream and downstream of that room: sustaining private, unwitnessed effort on a schedule nobody enforces, keeping the gap between the well-read and the merely present from splitting the group, and building — deliberately, at some cost to the circle's original informality — the institutional apparatus that lets an argument outlive the evening it happened in. Lenin's answer was an open hierarchy, Bakunin's a hidden one, the IRA's a structure that traded collective reasoning for survivability, the Birch Society's a discipline borrowed from the enemy and pointed backward instead of forward — four different prices paid for the same thing a book club needs far less desperately but still needs: some mechanism, however costly, for making sure the group outlives the room it met in, and stays honest about whether the room was ever right. Every reading circle that mattered eventually had to stop being only a reading circle. Most of them never made that choice, which is why we remember the ones that did — for better and, as often, for worse.

Can people be trusted as a social group?

Every structure in this essay is an answer, whether its builder said so or not, to a single underlying question: can the people in the room be trusted to hold a commitment without something external forcing them to? Lenin's answer was no, or not yet — trust had to be manufactured by professionalizing a cadre and making membership a job with standards, because he'd watched circle after circle prove that sincerity alone didn't survive an arrest or a hard winter. Bakunin wanted the answer to be yes and couldn't quite make himself believe it; the Invisible Dictatorship is what an anti-authoritarian builds when he trusts the movement's aims but not its members' collective judgment under pressure, and builds it secretly because admitting the distrust would have meant admitting the philosophy didn't scale. The IRA's cells answer the question by refusing to ask it of anyone below the top: no member has to be trusted with more than a handful of names, so no member's possible failure can cost more than a handful of names. The Birch Society trusted its chapters with a fixed diagnosis and very little else, which is its own quiet answer — sincerity, yes; judgment, no further than the material provided.

The committees of correspondence are the one case in this essay where the answer comes back yes without qualification, and the reason is worth sitting with, because it isn't that eighteenth-century Massachusetts produced more trustworthy people than nineteenth-century Russia or twentieth-century Ireland did. It's that the correspondence itself did the work trust is usually asked to do on faith. A committee's position was written down, dated, sent to a dozen other committees, and available to be checked against whatever that committee did next — which means nobody had to simply believe a given town would hold the line; they could see, in the next month's letters, whether it had. Trust, in other words, wasn't a precondition the committees required of each other before cooperating. It was a byproduct of a structure that made backsliding visible and comparatively costly to hide. Jefferson and Adams's own decades of correspondence work the same way in miniature: two men who had genuine reason to distrust each other's late-life politics kept writing anyway, because the letters themselves — read, kept, eventually published — were the mechanism that let each check the other's position against what he'd actually said, not what a rumor or a newspaper said he'd said.

That suggests the honest answer to whether people can be trusted as a social group is that it's the wrong axis of comparison. The kruzhki weren't full of less trustworthy people than the committees of correspondence were; they were full of comparably sincere people operating without any mechanism that turned a private wavering into a visible, checkable fact. Trust, on this accounting, is less a property of character to be sorted for at the door and more a property of design — of whether a structure makes commitments legible to the other people relying on them, or leaves each member's follow-through invisible until the moment it's tested and found wanting. A group that builds in nothing checkable is betting on virtue holding under pressure with no way to notice early when it doesn't; a group that builds in correspondence, minutes, dues records, a newspaper, anything that turns private intention into a public trail, doesn't need everyone to be especially virtuous, only checkable — which is a lower bar, and, on the evidence of two centuries of circles that didn't survive their own optimism about the first bar, a far more reliable one.