For roughly two years after the First World War, Italy looked to almost everyone living through it — workers, employers, and the government alike — like a country going through a revolution. Then it stopped, and within two years of stopping it produced the first fascist government in Europe. Those two facts sit next to each other awkwardly, and the whole point of this course is to work out how they connect.

One word you need from the start, in outline only. Italian fascism begins as a movement of ex-servicemen and nationalists around Benito Mussolini — who had been editor of the socialist daily Avanti before the war, and was expelled from the Socialist Party in 1914 for supporting Italian intervention. Its members organized in local bands called fasci — leagues — which is where the movement's name comes from.

The founding rally of the fasci was held on 23 March 1919, in Piazza San Sepolcro in Milan, and did not formally adopt a programme.

It announced itself three weeks later.

On 15 April 1919 nationalists and fascists burned the offices of Avanti, the Socialist Party's own daily paper, in Milan — and the PSI's response was supine: it relied on the forces of public order.

Hold both halves of that. The attack came in daylight, in the same city where the occupation would peak seventeen months on; and a party that would shortly be the strongest single force in parliament answered an assault on its own newspaper by calling the police. What later turned that small and doctrinally empty movement into a mass force was an armed squad movement in the provinces. Where it came from, who joined it, and how it won are the business of Part IV; for now, take fascism as the thing waiting at the other end of the story.

Italian historians call the postwar crisis of 1919-20 the biennio rosso, the Red Biennium.

The label is worth pausing on, because it names a period and quietly implies a verdict — that these were the years when the thing was possible. Whether that verdict is right is precisely what is in dispute, so treat the phrase as the name of a question rather than the answer to one.

Numbers first, because the scale is genuinely hard to believe if you only know the ending. Two organizations recur constantly from here on: the Italian Socialist Party — the Partito Socialista Italiano, the PSI — and the socialist trade-union confederation, the Confederazione Generale del Lavoro or CGL. Module 3 gives their full anatomy; for now, take them as the movement's political wing and its industrial wing. Here they are at their peak, in the winter of 1919 into 1920.

At the general election of November 1919 the PSI took 1,834,000 votes and 156 deputies in the Chamber, making it the strongest single force in parliament.

Note what 156 deputies meant: the largest bloc in a Chamber of some 500 seats, not a majority. The PSI could dominate no government, and that gap between electoral weight and governing power runs through the story.

Trade unionism reached 3,800,000 workers, five times the prewar total.

That 3.8 million is all organized labour in Italy, not the socialist share of it: there were Catholic and syndicalist union federations too, and Module 3 sorts them out. The socialist portion is the one that matters here.

The CGL had 250,000 members at the end of the war; within two years it had two million.

The PSI, 200,000 strong, took control of 2,800 communes — 24 per cent of the total.

Read that last figure slowly. A commune is the basic unit of Italian local government — a town or city and its territory, with an elected council and mayor. So this was not only a parliamentary bloc and a pile of union cards: a quarter of Italy's communes were already being administered by the movement. A movement of that size does not need to seize the state in one dramatic night to be a standing problem for the people who own the country.

The peak came in the late summer of 1920, out of an ordinary wage dispute in the metal industry. The employers refused to negotiate and the workers answered with a go-slow. Crucially, the metalworkers' federation — the Federazione Italiana Operai Metallurgici, FIOM, the most powerful industrial union in the country — had already decided what it would do if the employers retaliated by locking them out: rather than be shut outside the gates, the workers would stay at their machines and hold the plant. So when the lockout came at the end of August, the response was not improvisation but a prepared riposte — and the factories became theirs.

The occupiers numbered over 400,000, and the total reached half a million as the labour force of non-metallurgical plants joined them.

For most of a month, half a million workers held the plants of industrial Italy with arms available to them, ran red flags up the chimneys, and in many places kept production going under their own management. A reporter for Il Corriere della Sera walked past the Milan factories on the evening of 31 August and filed a description worth having in full, because it shows what "occupation" actually looked like on the ground.

The spark was the Romeo lockout in Milan: the metalworkers learned of their organization's decision to occupy during the lunch break, and prepared to stay overnight in the factories.

One reached the factories, he wrote, through crowds of women and children coming and going with dinners for the strikers — “voluntary prisoners of the factories.” The entrances were strictly guarded by groups of workers. “Not the ghost of an official or police officer in sight,” he noted; “the strikers were complete masters of the field.” Anyone passing in a car or cab was stopped and checked “as if he were crossing the frontier.”

Hold onto that last image — and then be careful with it. The workers were running border control on the public roads of Milan and no policeman was arguing. But the police were absent by order: prefects and chief constables were following precise instructions from a government that had decided not to contest the factories. An absent state is not the same thing as a collapsed one, and telling those two apart is the hinge of this whole course — Module 20 is where the government's reasoning gets taken apart.

On that first night in Milan the workers at the Breda plant locked up two civil engineers — the only detention the paper reported — and they were, in Il Corriere's own words, “treated with respect.”

This was not a riot.

The sense that the ground had shifted reached well beyond the left.

Luigi Albertini — editor of Il Corriere della Sera, Italy's most influential newspaper, a liberal and a senator — called for the socialists and the CGL to assume power; there was a widespread feeling that the transition to a socialist economy had begun.

Albertini held no power to give, so this is a newspaper editor's reading of where things stood, not a transfer of anything. But that is the point. When the liberal senator who edits the country's leading daily concludes that power should pass to the socialists, "revolutionary situation" stops being a phrase militants use about themselves and becomes a description that people outside the movement accept. That is the high-water mark. Everything after it goes the other way.

With half a million workers holding the factories, one question could no longer be postponed: was the occupation the opening move of a bid for power, or an asset to be traded for the best settlement on offer? In September 1920 the party and union leaderships gathered in Milan to answer it — and this is the hinge of the entire period, stranger than either a victory or a massacre would have been. The decision fell to the national council of the CGL — an assembly of union officials, delegated by the local chambers of labour (the camere del lavoro, each the hub of a town's unions) and by the trade federations, the largest of which was the land-workers' federation. Spriano's description of what that meant is worth keeping: the union council was erected into a kind of court of appeal for the socialist movement, summoned to pass judgment on a revolution proposed by the party. Then they divided, and counted.

The proposal was rejected by 591,245 votes to 409,569, with some 93,000 abstentions — many of them FIOM sections, the metalworkers whose members were occupying the factories.

Representatives of the two autonomous unions — the railwaymen and the maritime workers — had no vote at all.

The winning motion was moved by Ludovico D'Aragona, the CGL's own reformist secretary; the losing, revolutionary one was signed by Schiavello and Bucco of Bologna, and reflected the party's position.

Nearly half the CGL membership were labourers in the agricultural sector, and the leaders of the land-workers' federation rallied to D'Aragona because they were opposed to committing the rural workers to the struggle.

And the revolutionary syndicalists — the Unione Sindacale Italiana or USI, a rival union federation outside the CGL and the one wing of the movement certain to vote for insurrection — were not invited to that meeting at all. (Reformist and revolutionary syndicalist are the movement's own factional labels; Module 2 pins them down.)

Those totals are delegate votes cast on behalf of member organizations, not a head count of workers, so read them as the weight of the CGL's constituent bodies rather than a referendum. And notice what the vote was actually between. The winning motion was not "do nothing"; it was to direct the movement towards union control of industry — a negotiated transfer of authority inside the firm, conceded by employers and government, rather than a bid to take the state. The losing motion was the revolution.

Gwyn Williams, whose history of the council movement is one of this course's main sources, calls that decision an overt act of historic renunciation. His verdict is on the choice. The further claim — that referring insurrection to a vote at all had already settled it — is an inference worth examining on its own, and you are not obliged to accept it.

It is easy to write this period as an ascending workers' movement facing a panicking ruling class. The evidence for the employers' state of mind does not support that. Confindustria, the employers' federation, refused even to discuss the metalworkers' wage claim that summer, and its spokesman explained why in one sentence.

Confindustria's spokesman on the wage claim: "Since the war, we've done nothing but drop our pants. Now it's your turn."

They had grounds for the confidence: they had already broken the Turin factory-council movement with a lockout supported by 50,000 troops.

Read on its own, that is just an insult, and a cornered opponent can shout as loudly as a confident one. What makes it evidence is the reasoning behind it, which Williams sets out in the introduction he wrote for Spriano's study.

Confindustria was secure in its strength and in its expectation of an unemployment crisis — the employers expected a recession that would do their bargaining for them.

That is a forecast, not a mood, and it turns the refusal into a strategy: if unemployment is coming, every month of delay weakens the union. Any account of 1920 that treats the workers as nearly victorious has to explain why the other side was making plans on the assumption that it would win.

The occupation ended in a negotiated settlement in late September 1920 — the terms, and the argument over whether they amounted to a victory, are Module 22. What followed was not a plateau but a rout, and its speed is the fact this course exists to explain. One thing to carry forward: in January 1921 the PSI split, and its revolutionary wing walked out to found a separate Communist Party of Italy — some 40,000 militants at its founding. Module 28 tells that story; here it matters only because from 1921 there are two parties to count instead of one.

By the PSI's congress of October 1922 its membership had fallen to 25,000, and that of the new Communist Party to 24,000.

On 28 October 1922 the fascists marched on Rome and took power.

By the summer of 1922 the fascists had taken over every former proletarian stronghold in Italy except Turin and Parma.

That is the shape of the defeat in one line: not a single decisive battle but a town-by-town conquest of the places the movement had held, until two cities were left. Turin — the city of the factory councils, and of most of this course — was one of the last two standing.

Be fair to those figures: the two successor parties together held about 49,000 members, so the honest comparison is 200,000 in 1920 against roughly 49,000 two years later — still a collapse of three-quarters, and alongside it a fascist government. Whatever destroyed the Italian workers' movement did its work in about twenty-four months, and it did most of it after the moment everyone remembers.

One warning about that date. Adrian Lyttelton, whose study of the fascist takeover you will meet in Part IV, insists that the seizure of power did not begin or end with Mussolini's famous March on Rome — it was achieved by gradual subversion of the liberal order (the parliamentary system then governing Italy, the one the PSI's 156 deputies sat in), starting in 1921 and not concluded for years afterwards. So 28 October 1922 is a convenient marker, not an explanation. The work that mattered was done in the provinces, and this course will spend a whole part there.

The historians in your sources do not agree, and this course will not pretend they do. Three families of explanation run through everything ahead.

One: there was no revolutionary moment. The opportunity people mourn was never really there — the employers were strong, the machinery of the state was intact and merely being withheld (remember the police who were absent by order, not by collapse), and the occupation was a defensive movement misread as an offensive one. Paolo Spriano, whose study of the occupation these modules lean on heavily, closes it by asking whether September was really the revolutionary moment. That the question is still open in the standard account is itself the point: this is a live possibility, not a settled finding, and Module 23 is where you will decide.

Two: the leadership failed. The moment existed and the movement's own institutions refused it — a party formally committed to revolution that would not organize one, and a union leadership that preferred a negotiated settlement it could administer. (That formal commitment has a name, the maximum programme, and Module 2 explains why a party could hold it sincerely and still never act on it.) On this reading September 1920 is a decision, not a fate.

Three: the industrial working class was alone. The factories were never the decisive terrain. Italy was still overwhelmingly rural, the countryside had its own explosive and quite differently organized revolt, and the two never fused. Every clause of that is a claim to be tested rather than a fact you have been given — Part IV is where the evidence sits, and it is the part most courses on this period leave out. Notice, though, that the September vote was taken by an assembly of union delegates in which the rural federations were the largest bloc.

The fascist offensive was to reveal, in all its stark horror, the isolation of the industrial working class and the factory proletariat at its heart — and that isolation registered first and foremost within the socialist movement itself.

Hold all three at once — they are not mutually exclusive, and the most defensible position at the end may well combine them. But each predicts something different about what you will find in the evidence, and you are about to go and look. Modules 2 and 3 give you the vocabulary and the institutions first, because almost every argument ahead turns on distinctions this movement made about itself.

One map before you start, because the parts ahead get referred to by name long before you reach them.

Part I · The ground (1889–1919), Modules 1–6. Who this movement was, what it called itself, what machinery it had, what the war did to it, and the year it won more than in its whole previous history without getting any closer to power.

Part II · The councils (1919–April 1920), Modules 7–16. One institution in one city. The factory council, from the shop-floor committee it grew out of, through the theory built on it and the two critiques from the left, to the lock-out that broke it in April 1920. This part is narrow on purpose.

Part III · The occupation (May–September 1920), Modules 17–23. The climax this module opened with: half a million workers holding the plants, the government deciding not to contest them, and the vote. It ends on the question the three explanations were built for — whether a revolution was available.

Part IV · The counter-revolution (1920–22), Modules 24–31. What came after, and mostly not in the factories: the agrarian offensive in the Po valley, the squads, the collapse of the movement's local power, and the March on Rome. If you take one thing from the shape of this list, take that the last part is the longest and is largely about the countryside.

Parts III and IV are still being written; where a module ahead is named and you go looking for it, that is why.

Module 1 left three words hanging: maximum programme, reformist, revolutionary syndicalist. This module pays them off and adds the two that do the most work in this course, maximalism and functional centrism. That may look like throat-clearing before the real history, and it is not — almost every argument ahead is conducted in these terms, and the September 1920 vote you have already seen is unintelligible without them.

One warning before we start. These are not neutral analytical categories invented by historians. They are the labels the movement used about itself, in public, as weapons. People adopted them, disowned them, and accused each other of secretly being something else. So the meanings move, and where a word's meaning drifted, the drift is usually the most interesting fact about it.

Start with the device that made all the other positions possible. A socialist party in this period faced an awkward question: it wanted to abolish capitalism, but in the meantime it was contesting elections, running town halls, and negotiating wages under capitalism. What is the relationship between the two activities?

The PSI's answer, adopted in 1900, was to hold two programmes at once: a maximum programme and a minimum programme.

The maximum programme is the socialist revolution — the abolition of capitalist property and the taking of power by the working class. The minimum programme is the list of things worth winning before that day comes: the vote, the eight-hour day, factory legislation, municipal services.

Held honestly, this is not a contradiction; it is a division of labour across time. The trouble is that it is silent on the one question that matters in a crisis — when does the minimum give way to the maximum, and who decides? Nothing in the structure answers that. Keep the gap in view: in September 1920 a movement holding both programmes had to choose between them under pressure.

One body was supposed to be the guardian of the maximum programme: the PSI's Directorate, the party's national executive, elected by its congress. Remember it — almost everything in this module ends up being about what that body did and did not do.

The function of the PSI's Directorate was to preserve the maximum programme, that is, the socialist revolution.

The reformists took the minimum programme seriously and the maximum programme as a distant horizon. Socialism would arrive through accumulated conquests — suffrage, legislation, union contracts, co-operatives, socialist municipalities — each one real, none of them insurrectionary. Their strength lay exactly where that strategy pays: in parliament, in the unions, and in the communes.

Their outstanding figure, and the man who effectively founded this tradition in Italy, was Filippo Turati — a Milanese socialist, later a deputy, and the reformists' voice for the next thirty years. You will meet him again at the end, still arguing for the same strategy in 1922.

Turati broke with the democrats of Milan and formed a socialist league there in 1889, the year the Second International was founded.

Turati launched Critica Sociale in 1891, the journal of reformist socialism and one of the great journals of social democracy — the broad European current that sought socialism through parliamentary and trade-union means.

Two things about the reformists are easy to get wrong, and both matter later. They were not a small right-wing faction: they ran the union confederation and dominated the parliamentary group, which is to say they controlled the movement's two most powerful machines. And they were not pretending: by 1914 accumulation had built something substantial, and they had the returns to prove it.

By 1914 the socialist movement had 52 deputies in the Chamber and commanded a million votes; it had won control of 300 communes and four provinces, and built up a whole apparatus of co-operatives, credit institutions and defence organizations.

Against them stood everyone committed to re-affirming the maximum programme. That is the original meaning of maximalist: not a specific faction, but anyone insisting the revolution stay on the agenda. In the postwar years the word narrowed to a particular stance — a loud commitment to revolution and to the dictatorship of the proletariat (the Marxist term for the period of working-class state power that follows a revolution), and a refusal of formal compromise with bourgeois institutions: that is, the parliament, monarchy and ministries of the existing propertied order.

Then it drifted again. Williams records that in practice the conduct of much of the party — above all the parliamentary group and the union leadership — was reformist, or else “void of any revolutionary content except the verbal”, and he is explicit that the word itself followed the conduct: maximalist, he writes, tended to acquire that meaning as well.

So maximalist ended up naming a commitment to revolution that was verbal — a commitment real as rhetoric.

Symbolism came to carry excessive weight as a defence against betrayal: an audience with the king was the kiss of death, and Bissolati — the reformist leader who accepted patriotic affiliation — was expelled even though he wore a soft hat and lounge suit to restore his socialist virginity.

That is a devastating thing for a word to come to mean, and you should hold it sceptically rather than swallow it: it is one historian's judgement, and the maximalists themselves would have rejected it. But keep it as a hypothesis with a test attached. If the charge is right, then a leadership of this kind, facing a real chance of revolution, will re-affirm the maximum programme and decline to act on it. The last card of this module is the first test; Module 21 is the full one.

Maximalism was a position, not an organization. The organized form it took before the war called itself intransigent revolutionary — a fraction, meaning an organized tendency operating inside the PSI rather than a separate party. Intransigent here means refusing all alliance with bourgeois parties: no electoral pacts, no support for liberal governments, no ministries. So intransigents are a subset of maximalists, distinguished by what they refuse rather than what they promise.

The fraction formed in October 1910, to resist the PSI's drift towards Giolitti — the liberal prime minister who dominated Italian politics before the war.

It got its journal's name from one of his insults. In April 1911 Giolitti remarked that Karl Marx had been “banished to the attic” — filed away upstairs with the other obsolete furniture, now that socialists were respectable enough to be governed with.

The intransigents answered on 1 May 1911 by launching a journal called La Soffitta — The Attic — to re-affirm marxism.

Naming your paper after the insult is a good joke, and it tells you what the fight was about: whether Marxism was a live instruction or a museum piece.

The movement spread: it won two deputies, 200 party sections — the PSI's local branches — and eight weeklies.

But Williams adds a dry qualification worth carrying: it was little concerned with marxism in fact, and functioned as a spokesman for activism. The paper defended Marx's name; what it actually supplied was militancy, which is not the same commodity.

The fourth position rejects the terms of the other three. The revolutionary syndicalists held that parties, parliaments and programmes were beside the point, and that workers should fight where they actually have power — at work.

They scorned politics and political parties, believing the class war should be fought on class terrain, that is, within industry or on the land.

The positive programme followed from that. Build industrial unions strong enough to constitute an anti-state — a parallel structure that could run production without the existing government — and then take over industry through the expropriating general strike. The mechanism is worth spelling out, because the phrase hides it: if every worker stops simultaneously and the unions are already organized to restart production on their own account, ownership becomes irrelevant. The owners keep the deeds and lose the factories. They prized spontaneity and distrusted bureaucracy, including their own.

This is the tradition behind the USI, which Module 1 introduced as the federation excluded from the September vote — and behind the anarchists, a related but distinct current. Take the difference now, because both act in Part I. Syndicalists have a strategy for taking over production: build unions strong enough to run industry, then take it with the expropriating general strike. Anarchists reject the state as such rather than proposing to capture or replace it, and are correspondingly readier to treat a spontaneous rising as the thing itself rather than as material for an instrument. The two overlap in practice — the USI's own secretary was an anarchist — and Module 14 takes the relationship apart. For now, note the shape of the disagreement. Reformists and maximalists argue about when to make the revolution, and intransigents about whom you may deal with in the meantime — all three accepting that the party is the instrument. Syndicalists reject the instrument: they argue about where the fight belongs.

The syndicalists set up a headquarters in Parma and in 1912 founded their own federation, USI, with about 100,000 members, concentrated among the lower-paid workers of Emilia; its objective was the expropriating general strike.

One term left, and it is the one that does the most work in this course. A party containing all four positions has to hold together somehow. The mechanism it evolved has a name.

Charged with preserving the maximum programme while its parliamentarians and unions pulled towards reform, the Directorate was trapped in a permanent functional centrism.

Take the phrase apart. Centrism because the Directorate sat between the two poles this module has been describing — the reformists with their parliamentary and union machines on one side, the maximalists and intransigents on the other. Functional because it was not a doctrine anybody argued for; it was the behaviour the party's structure produced, whoever happened to be sitting there.

Williams describes the cycle. Discontent breaks out among ordinary people; the reformists respond in democratic terms; the left protests in marxist terms; and the Directorate answers by re-affirming the maximum and revolutionary programme. The re-affirmation is the point — it satisfies the left rhetorically without requiring the reformists to do anything differently on Monday, so the party survives another round intact.

In the middle of the Giolittian decade socialist party membership was about 70 per cent working class, while its parliamentary representation and even its local leadership was 80 per cent middle class — a leadership with, as Williams puts it, precious little direct contact with working people.

Read that as a machine and its output is predictable: it converts pressure into re-affirmations. That is a structural property, not a moral failing, and it makes a testable prediction — a party built this way will restate the maximum programme whenever it is pushed, and will not convert the restatement into an instruction it issues in its own name.

September 1920 is the first test of that prediction, and Module 1 showed you only the end of it. Over 9–11 September the party and union leaderships met in Milan. Note who was holding the reins at that moment: the occupation was an industrial dispute, so under the jurisdiction Module 3 describes it was the CGL that had called the national council, the CGL that commanded the strike machinery, and therefore the CGL that was directing an occupation of several hundred plants. That command is the thing it now put on the table.

The CGL offered to resign the direction of the movement to the PSI Directorate if the party would propose to make the revolution — and the Directorate refused the offer.

So the offer was not a resignation of union office or a gesture of protest. It was an offer to hand over operational command of the occupation itself, on one condition: that the party declare for revolution. Read carefully, because it is easy to misread as the party disowning the revolution, and it is not. The party's position still went to the union council — as the motion Module 1 called Bucco's, which Spriano records was signed by Schiavello and Bucco of Bologna and reflected the party's position.

D'Aragona, the CGL's secretary, told delegates they should be absolutely clear that a vote for the Schiavello-Bucco resolution was a vote for communist revolution and the dictatorship of the proletariat — and he expressly rejected the Russian model for Italy.

And here is what that motion actually asked for. It requested “the directorate of the party to take over the direction of the movement and to lead it towards the maximum solution of the socialist programme.” So the revolutionary motion at the union council was a motion asking the party to do the thing the party had just declined to do when asked directly.

That is functional centrism in its purest observable form. The maximum programme is re-affirmed — in a resolution, at length, in the correct words. What never appears is an instruction issued by the body constitutionally charged with the maximum programme. Asked to lead, it declined; asked again by resolution, it was outvoted, 591,245 to 409,569. Spriano adds the detail that completes the picture: the party's own national council — the wider body above the Directorate — was never assembled at all, because it could not be summoned in time.

One of the revolutionary motion's own supporters, Bensi of Milan, said afterwards what the procedure looked like from inside: “a revolution is not made by first calling a convention to decide whether there is going to be a revolution or not.” That is the sharpest one-sentence statement of this module's thesis, and it comes from a man who voted to make the revolution.

A caution before you bank it. This explains the party's conduct; it does not yet explain why a union body was entitled to decide a question of revolution at all. That is an institutional question, not a doctrinal one, and Module 3 answers it. The two explanations are layers, not rivals.

Module 2 gave you the positions people held. This module gives you the machines they held them in — and the machines matter more, because in September 1920 the outcome was shaped less by what anyone believed than by which organization had the right to decide.

The single most useful thing to understand about the Italian workers' movement is that it was not one organization. It was several, with separate memberships, separate leaderships, separate money, and a written agreement about who was allowed to do what. Keep asking, of every episode in this course: which machine is this, and what is it entitled to do?

Start at the bottom, with the institution an English-speaking reader is most likely to underestimate. The camera del lavoro — chamber of labour — was the hub of all the workers' organizations of a town or district: local unions, leagues, co-operatives, savings banks.

The camera served initially as a worker-controlled labour exchange and information centre to improve the bargaining position of labour, and grew from there into the real centre of working-class life.

A labour exchange here means something with teeth. Where the camera was strong enough, employers had to come to its register for workers instead of hiring at the gate or picking a gang out of the piazza each morning. Corner gives the clearest statement of the ambition, from Ferrara: what the uffici di collocamento — labour exchanges responsible for the distribution of work — were for was a monopoly of labour in the province, making the landowners dependent on the workers' own organization. That is also how a wage agreement gets enforced without a strike: an employer who breaks the rate finds that nobody is sent to him.

Now notice where such a power can bite, because it shapes everything that follows. A rotating register only works where labour is taken on by the day or the season, in gangs — the braccianti of the Po valley above all, and in the towns the building sites, the docks and the carters. It has nothing to grip in a Turin metal plant, where the workforce is settled enough to elect standing committees out of its own shops — as the modules on the factory will show you it did. So the movement's two halves reached for two different weapons: in the countryside the register, and in the factory the workshop itself — which is the institution the whole of Part II is about.

Corner adds the sting: before the war the Ferrarese monopoly was never actually obtained, and strikes were correspondingly leaky. Winning it was the great postwar prize in the countryside, and Part IV is where the fight over it decides the outcome of this story.

The first camera del lavoro was founded in Milan in 1889, in imitation of French labour organizations of syndicalist inspiration.

Williams warns against the obvious English analogy: the camera was “far more significant locally than its formal British equivalent, the trades and labour council.” Because it could mobilize a whole district behind one group of workers, the local general strike became a standard feature of Italian labour practice — and the camera, being closer to popular feeling than any craft federation, was a natural theatre of radical action. Remember this when Turin's camera votes in Module 12.

The camera's headquarters was the casa del popolo, the people's house: hiring hall, labour exchange, school, club, meeting place. For a working-class family it was where you looked for work, got an education, had an argument and spent an evening.

The casa del popolo was a prime target for the fascist squads during 1921-22, and its destruction could cripple the local movement.

That sentence is worth stopping on, because it explains a tactic you will watch in Part IV. Squads did not need to defeat the movement in an argument or an election. Burn the building and you have removed the labour exchange, the meeting hall and the school at one stroke — the physical plant of working-class life in that town. The movement's local strength and its local vulnerability were the same address.

The CGL — Confederazione Generale del Lavoro — was founded in 1906, in reaction against syndicalism, and was the nearest Italian equivalent to the British Trades Union Congress.

It federated craft and industrial unions together with the camere, rising through a national council to a small directive council. Note one structural detail with consequences: the directive council was not elected directly by congress but recruited from the major unions, which insulated the leadership from the membership's mood. Its leaders were generally reformist socialists and its journal was Battaglie Sindacali, Union Struggle.

Hold its two million against the party's 200,000: the union was ten times the party. But the revolution was constitutionally the party's business. The Directorate existed to preserve the maximum programme, whereas the CGL was founded in reaction against revolutionary syndicalism, was led by reformists, and — under a treaty with the party that this module's card on the Pact of Alliance takes apart — was confined to economic action, meaning wages and conditions rather than political power.

Be careful what that does and does not mean. It is not a claim that the union's members wanted no revolution — in September 1920 over 400,000 votes in the CGL's own council were cast for making one, and the motion that carried those votes asked the party's Directorate to take over the direction of the movement. Even the union's revolutionaries treated the revolution as the party's job. The claim is about where the mandate sat and where the power sat, and that they were not the same place.

Inside the CGL, the largest single body was not industrial at all.

The land-workers' federation, Federterra, reached the 900,000 mark in 1920 — the strongest single grouping inside the CGL.

These land-workers have a name you will need through Part IV: the braccianti — literally "arm-men", landless hired labourers, paid by the day. On the Lower Po, Williams writes, farming had become industrial, with its teeming braccianti organizing in tough unions against the tenant farmers and the big agrarian owners. They are a rural proletariat — wage workers who happen to work fields instead of lathes — which is why they belonged in a socialist labour confederation at all.

And they were one class among three. Williams is blunt about the geometry: the division between socialist braccianti and catholic mezzadri — sharecroppers — and small proprietors was structural.

Hold that three-way map: it is the countryside in one sentence. The braccianti were socialist and in Federterra; the mezzadri leaned Catholic, toward the CIL and the PPI you will meet in the card on the rest of the field; the small proprietors — whose numbers the war swelled, as Module 4 will show — owned the thing socialism proposed to abolish. Nine hundred thousand of the CGL's two million were braccianti. Set that beside the roughly 160,000 metalworkers and the picture of the movement changes shape: the factories that gave the Biennio Rosso its images were a minority interest inside their own confederation. This is the single most important number in Part I. When Module 1 told you the September assembly was one in which the rural federations were the largest bloc, this is why — and it is the institutional form of explanation three, the isolation of the industrial working class.

FIOM — the Federazione Italiana Operai Metallurgici — was founded in 1901 with its headquarters in Turin, and claimed about 160,000 members at its postwar peak.

Module 1 already told you FIOM was the most powerful industrial union in the country. What matters here is who ran it: Williams notes it was led by the able Bruno Buozzi — a reformist, and a competent one, which matters because much of Part III turns on decisions he took. Being based in Turin also makes FIOM the union that the factory council movement will later try to capture from below: the councils are the subject of Module 9, and until then treat the phrase as a placeholder.

The PSI was built upward from urban sections — local branches — through provincial federations to a national council and the Directorate you met in Module 2, elected by party congress.

The real leader of the mass party was generally the editor of its journal Avanti, founded in 1896 and moved to Milan in 1912.

Two components sat awkwardly inside that structure, and both come back later. The parliamentary group — the Gruppo Parlamentare Socialista or GPS — was nominally subject to the Directorate but in practice enjoyed an autonomous existence, and was the reformists' stronghold. And the youth federation was autonomous too, with its own journal Avanguardia; it is where several of this course's central figures start out.

So the party that was constitutionally committed to revolution contained a self-governing bloc of parliamentarians committed to reform. Module 2 called the result functional centrism. Here is its wiring diagram.

Now the card that makes September 1920 intelligible. Party and union were allies, not branches of one body, and they had divided the world between them in writing.

The CGL claimed exclusive control over economic as opposed to political strikes, a division defined at the Stuttgart congress of the Second International in 1907 and confirmed in the Pact of Alliance signed with the party in September 1918.

Read that as a constitution and the trouble is immediate. A strike over wages belongs to the union. A strike over politics belongs to the party. But the whole content of a revolutionary crisis is a dispute that is both at once — half a million workers occupying factories over a pay claim, which is also a question about who owns industry and who governs Italy.

The Pact does not adjudicate that case; it simply hands the same event to two owners. So on paper the crisis is unanswerable — but paper was not what settled it. The union was ten times larger and held the strike machinery, so in any contest over a stoppage the practical answer to "who decides" was fixed in advance by weight rather than by the text. When the CGL's national council voted on the revolution it was not usurping anything; it was exercising a jurisdiction the party had signed away two years earlier and could not have enforced in any case.

Keep both halves, because they are doing different work and Part III will test them against each other. The constitutional half says nobody was entitled to decide. The practical half says the union was always going to. If the second is right, the Directorate's refusal in Module 2 changed less than it appears to; if the first is right, that refusal was the moment the question went unanswered.

Finally, the movement was not alone on the ground, and three of its rivals matter.

The syndicalist USI — Unione Sindacale Italiana, founded in 1912, its postwar secretary the anarchist Armando Borghi — was claiming 800,000 members in 1920, with the journal Guerra di Classe.

The Catholic union federation — the Confederazione Italiana del Lavoro, CIL — was recognized by the Papacy in March 1918 and grew past a million after the war — essentially rural, politically ambiguous, and allied to the PPI, the Partito Popolare Italiano, the Catholic party founded by Luigi Sturzo in 1919 which took over 100 seats in the November election. So the countryside was contested ground, not socialist territory.

The UIL — Unione Italiana del Lavoro, the Italian union of labour, broke away from the syndicalists during the war crisis and was run by nationalist syndicalists, republicans and radicals who had supported Italian intervention.

Much smaller than the other federations, the UIL's left nationalism supplied recruits to fascism.

One more machine, on the opposite side of the table, and it was organized before the workers' side had finished arguing about jurisdiction.

Confindustria, the employers' confederation, was first formed in 1910 in imitation of the Turin Industrial League and revived in March 1920 — a very powerful and aggressive organization.

The date is not decoration. The employers re-formed their national body in March 1920, six months before the occupation, gathering regional and sectional consortia including the metal-industry employers who would refuse FIOM's claim that summer. Module 1 quoted their spokesman refusing that claim outright. That was a federation speaking with one voice, against a movement whose two halves had a treaty about which of them was allowed to answer.

Everything in Modules 1 to 3 — the two million union members, the occupied plants, the rival machines — sits on top of a transformation that took four years. Italy went into the First World War a mostly agricultural country with a modest industrial base, and came out of it with a heavy industry the size of a great power's and an industrial workforce swollen at exactly the places that would matter. Be precise about where the doubling happened, because the course will lean on it: Clark's estimate is that the number of industrial workers in Turin had probably doubled since 1913, in a city where nearly a third of the population was by then involved in industry. That is a claim about one city, not about Italy.

During the war Fiat's labour force grew from 7,000 workers to 30,000, and the worker population of Turin doubled.

Martin Clark, whose study of the factory councils is the fourth of this course's main sources, puts the claim at its strongest: the factory council movement, like its counterparts elsewhere in Europe, was a product of the First World War. So this module is not background. The war built the movement that reached September 1920 — and, in the same motion, built the force that destroyed it. Whether the first ever came close to a revolution is the question Module 1 told you not to prejudge, and nothing here settles it.

The economics were simple and enormous: a guaranteed market at guaranteed prices, paid for by the state, with loose contracting and very large profits. Expansion was fastest after Caporetto — the catastrophic military defeat of late 1917 in which the Italian front collapsed, and which produced both a national emergency and a frantic rearmament. Williams calls the growth that followed breakneck and almost blind.

Italian industry produced six times as many pieces of artillery during the war as the Italian army had possessed when it started.

On Armistice Day there were 7,709 cannons in the field — more than the British had. Aircraft production started virtually from scratch and built 6,523 planes in 1918 alone.

The money figures need one caution before you read them. Company capital across Italian industry rose by an average of 56 per cent, and in engineering by 152 per cent — but these are nominal lire, and as the card on the countryside's war will show, industrial prices roughly trebled over the same years. Against a trebling of prices a 56 per cent nominal rise is a real fall, so the all-industry average is not evidence of a boom at all. What is not inflation is the concentration at the top, where a handful of firms grew by an order of magnitude more than the average.

Ilva's capital went from 30 million lire to 300 million, Breda's from 14 to 114 million. And the biggest of them acquired a nickname, from Williams.

The colossus was Ansaldo, “the monster with a thousand gullets” — a Ligurian shipbuilder whose capital mushroomed from 30 million to 500 million lire and which employed 110,000 workers in its myriad plants.

In Turin, Fiat's capital rose from 17 to 200 million lire, and the city became an industrial monolith.

Now the part that matters more than the tonnage. To run a war economy the state did not merely place orders; it placed the factories themselves under a regime called Industrial Mobilisation, which applied to most large firms. A plant inside it was an auxiliary firm, and its workers, in Clark's words, were “subject to quasi-military discipline.”

Within the 1,976 auxiliary firms, workers were divided into four separate groups.

The groups are worth learning, because they are a map of who owed what to whom. Military or commanded workers were men actually in the armed forces, seconded to munitions factories — soldiers at a lathe, punishable under military law. Exempted workers would have been in the services but for a skilled trade. Civilian workers had no military liability. And fourth, women and youths under eighteen.

There were 905,000 workers in auxiliary industry, and the proportions matter more than almost anything else in this module.

Of the 905,000, only 331,000 — 36 per cent — were either "military" or "exempted" workers; the remaining 574,000 were 304,000 men without military obligations, 196,000 women, 60,000 youths and 14,000 prisoners of war or colonial workers.

Prato splits that 36 per cent again: roughly 17 per cent commanded and 19 per cent exempted. So the men the countryside would come to resent were about a third of the auxiliary workforce, and half of even that third were soldiers who had been put in a factory rather than men who had escaped the call. Keep those numbers; the card on the two mobilized masses turns on them.

Dismissal became the most effective punishment of all: for military or exempted men it could mean being sent to the front, and for everyone else it meant exclusion from all other auxiliary factories and the necessity of working for lower wages in some small workshop.

For that third, the categories were not paperwork. Leaving or striking was a military offence rather than an industrial act, and dismissal — for the military or exempted worker — could mean being sent to the front. The factory had the trench standing behind it as a punishment.

In the ten months from January to October 1918, 19,018 workers were sentenced to terms of hard labour, and a further 9,522 to spells in an ordinary prison.

Resist reading those as a measure of mass punishment — Clark's own footnote calls them a tiny fraction of the wartime labour force. They measure deterrence. The measures were "undoubtedly effective in enforcing discipline", he writes; absences fell sharply, and in the whole war the only important agitation was the Turin rioting of August 1917. A threat does not have to be used often to be believed.

Which is the sting in it. The discipline worked, and Clark's verdict is that it worked at a price: imposed on people who believed they were labouring in intolerable conditions merely to swell their employers' profits, it "seriously exacerbated class hatred in the big industrial cities". Hold that against Module 3's picture of a movement arguing over jurisdiction. In the plants that mattered most to the war, a substantial part of the workforce had no industrial rights to divide — and was being given a reason to want more than rights.

Industrial Mobilisation also had a machinery, which Module 7 will measure against what was actually happening on the shop floor.

Industrial Mobilisation rested on a Central Committee in Rome, where the leading arms manufacturers sat — Agnelli of Fiat, Breda, Perrone of Ansaldo, Caproni — and on Regional Committees made up of a high-ranking army or navy officer as President, two representatives of the employers, two of the workers, and two civilian economic experts.

Note the shape of that committee, because Part II turns on it: a dispute in a war plant went to a body chaired by an officer, with employers and workers sitting in equal and small numbers.

What that regime felt like can be given in one number.

A 70-hour week was compulsory at Fiat by March 1916, and nearly all workers in fact put in 75.

Seventy-five hours is six twelve-hour days with time over. Clark notes the consequence that turned long hours into political trouble: they fell hardest on women, who now had both the shifts and the shopping, in cities where public transport and food were scarce. Clark records that discontent focused on food and housing rather than on pay — scarcity as much as price — and it was food that put Turin into the streets in August 1917, which is Module 5.

The other Italy was being mobilized far more brutally at the same time, and this is the half that most accounts of the Biennio Rosso skip.

Of the nearly six million men conscripted, 46 per cent were peasants or rural workers; 1,300,000 were killed or disabled.

Resist the obvious inference from that 46 per cent, because the figures do not support it. Williams estimates 57 per cent of Italy's population as living on the land in 1911, against those 46 per cent of conscripts — so on the only comparison the sources allow, the countryside was not drafted beyond its share. (Treat this as suggestive rather than settled: a share of the whole population is not a share of military-age men, and rural households were larger.) No source anywhere gives a rural-versus-urban split of the dead.

What the countryside did lose can be measured directly, and it is staggering.

The rural labour force of males over 18 fell from 4.8 million to 2.2 million — some 2.6 million men, better than half the adult male countryside.

That is what the war took out of rural Italy, and it squares with the conscription figures: 46 per cent of 5,750,000 drafted is about 2.6 million. The countryside also got promises. After the Caporetto disaster of 1917 — a military collapse so severe it triggered a mass rally to national defence — the government was lavish in its promises of land reform, because it needed peasants to keep fighting.

Agricultural prices doubled while industrial prices trebled.

And a real change of property followed: in northern Italy the proportion of peasant proprietors rose from 26 to 36 per cent as households fought through to some security from debt and mortgage. So the war left the countryside mobilized, armed with promises, and — for a substantial minority — newly owning something. Keep that last point; a peasant who has just acquired land is not obviously a recruit for socialization of the means of production.

Now put the two halves of the module together, and put them against Module 3. That module gave you Part I's single most important number — the 900,000 braccianti who made the movement's largest bloc a rural one. Module 3 showed you that isolation as an institutional fact: who could outvote whom inside the CGL. This module shows you the same isolation as a social one, and the two are not the same thing. A movement can survive being outvoted by its allies. It is in much worse trouble when the people doing the outvoting have a grievance against it.

The industrial working class grew and concentrated in a few northern cities and spent the war under quasi-military discipline in the plants. The countryside lost more than half its adult men. And the resentment between them is attested directly: Williams says it proved easy to whip up peasant hatred against exempted industrial workers, and Clark that the peasant infantry felt great bitterness towards them.

Now the fact that makes this module worth the trouble. That resentment ran on a belief, and the belief was false.

The proportion of exempted workers was not high, Clark writes, and the popular impression that all armaments workers were imboscati — shirkers, literally "hidden in the woods" — was false.

Set the two halves of the arithmetic side by side. The peasantry was not drafted beyond its share of the population — 57 per cent on the land, 46 per cent of conscripts. And the industrial workforce was not shielded as a class: barely a third of auxiliary workers had any military status at all, and half of those were soldiers ordered into a factory rather than men who had dodged the call. The antagonism that shaped the next five years was not a response to a statistical injustice. It was a response to something more dangerous: a protection that was visible, written in law and attached to an occupation, so that anyone could point at it and tell a story about it.

That is worth carrying into Part IV, where the story gets an organizer. A grievance founded on a false impression does not become weaker for being false — Williams's verb is that the hatred was easy to whip up, which is what you do to a belief rather than to a fact.

The first consequence of the war's two mobilizations was an antagonism. The second cuts against the revolution from the opposite direction, and it concerns what the industrial working class had not learned while it was at the lathe.

Urban workers had virtually no combat experience, and were imbued with an anti-militarist spirit that prevented them from making any serious attempts to secure arms or form military units, even after the war.

That sits awkwardly against Module 1's image of half a million workers holding the factories "with arms available to them", so resolve it now rather than carrying a contradiction. Both are true, and the difference between them is the whole point. Spriano's occupiers did have weapons to hand in the plants; what Clark denies is something else — military units, training, and any organized effort to acquire arms as a force. A pistol in a factory and an armed body capable of taking a city are not the same possession, and it is the second that a revolution needs.

Note also that this cuts against Clark in one place. Module 5 will show Turin workers in 1917 attacking barracks for arms — which is an organized attempt to get them. So "no serious attempts" is a claim about the postwar years, and you should hold it as Clark's judgement rather than as a settled fact. His verdict on the 1920 demobilisation is blunter still: the movement would be faced by a hostile army consisting largely of peasants.

Four things, and the rest of Part I is the working out of them. An industrial working class swollen and concentrated — roughly doubled in Turin itself. A cohort of that class trained in obedience under military discipline — and, as Module 7 will show, trained in the grievance machinery that discipline required. A countryside that had paid for the war in blood, been promised land, and partly acquired it. And between the two, a resentment.

The conventional summary of 1919–20 is that the war radicalized Italy. That is true and much too simple. The war radicalized two Italies, on different timetables, with different enemies, and left them suspicious of each other. Module 1 called the isolation of the industrial working class the third explanation for the defeat. This module is where that isolation was manufactured.

Three years before the occupation, the city that would become the centre of this course rose, fought the army in its own streets for a week, and lost. Almost nothing about the Biennio Rosso makes sense without it — not Turin's reputation, not Gramsci's career, and not the confidence with which people in 1920 assumed that Turin would move first.

It is also the cleanest natural experiment in Part I. In August 1917 an Italian city went into insurrection without any organization deciding that it should. What happened next tells you what the movement's institutions were for — and what they were not.

Module 4 left Turin working long weeks in plants under military discipline, with food and transport scarce. What follows is Williams's narrative of the week, which is fuller than Clark's and — as the card on choosing a test will show — not neutral between them.

On the morning of 21 August 1917, eighty bakeries failed to open in yet another breakdown in bread supplies.

The streets filled with protesting women and children. Wheat was rushed in, too late. Then, in Williams's words, the factories exploded into an anti-war demonstration — which is already more than a bread riot. Clashes between police and workers gave way to pitched battles between workers and soldiers.

The working-class quarters were barricaded, barracks attacked for arms, two churches sacked.

The churches are worth a moment, because "two churches sacked" invites you to supply anticlerical doctrine and the reason on the ground was smaller and older than that.

On 23 August anti-clerical incidents dominated the rioting: the church of San Bernardino in Borgo San Paolo was burnt down and its monks had to flee for their lives, after a crowd arrived to settle accounts arising from an incident in 1916 when the monks flogged several young boys caught stealing pears from their garden.

At Madonna di Campagna the church of La Nostra Signora della Pace was sacked and a red flag planted in its bell tower. Keep both halves of that: the flag says the rising read itself as a revolution, and the pears say the crowd was also settling a score from the year before last. A rising can be both, and the historians in the next cards divide over which half to weigh.

Hold that middle clause; it decides a good deal later. On 24 August the workers advanced on the centre of the city and nearly broke through, and were finally beaten back by machine-guns and tanks. By the 28th it was over. At least fifty workers had been killed and several hundred wounded.

The detail that matters most for this course is not that the army suppressed the rising. It is who the army was.

Sardinian peasants in uniform had no hesitation in shooting down Turin workers in August 1917.

Module 4 argued that the war mobilized two masses and left one resenting the other — and that the resentment ran on a belief that was false in both directions: the countryside was not drafted beyond its share, and the industrial workforce was not shielded as a class. Now hold the limit of what this street shows about that. It does not show these particular soldiers firing out of the imboscati grievance; no source here gives their motive, and men fire because they are ordered to. What it shows is the state reaching for one mobilized mass to put down the other, and finding it ready to hand — three years before the occupation, and five before the March on Rome.

Keep the shape of it. The industrial working class rose in the one city where it was strongest, and the state suppressed it using the other mobilized mass. No account of 1920 that treats the countryside as a neutral background can survive August 1917.

Now the fact that makes this a test of the party rather than of the city — and both historians agree on it, which is worth noting because they agree on little else here.

Clark finds the rioting completely spontaneous — no evidence of organisation by the local or national Socialist Party, or by any other group; Williams likewise says the movement had risen spontaneously from below.

But Williams adds a qualification that Clark's account lacks, and it matters: the immediate uprising was spontaneous, “but the militants of the section, anarchists and syndicalists, quickly moved in to organize defence.” So the question is not simply organized or unorganized. It is that the organizing was done by people the party did not control, after an event the party did not foresee.

On 23 August some thirty socialist and union leaders, intransigent and reformist together, hurriedly met to give some order to the movement. Williams reports the outcome in five words: “No-one knew what to do.” They issued a statement praising the workers' courage and asking them to give up any further useless violence; on the 26th socialist deputies asked the workers to go back to their jobs.

Meanwhile Giacinto Menotti Serrati — a maximalist, editor of Avanti and by Module 3's account therefore the effective leader of the party — travelled post-haste from Milan to Turin, “arriving just in time to be arrested as an instigator”. Over 800 socialists and anarchists were taken, including the whole of the Turin section leadership, and 200 were shipped off to the front. Williams's verdict is that apart from Serrati's action the PSI had shown no sign of life during the rising, and that it evoked no response anywhere else in Italy.

Clark's summary of the party's position: the Socialists were not responsible for the rising, but they felt they ought to be. Read that beside Module 2. Functional centrism described a party that converts pressure into re-affirmations; here is the same party meeting in the middle of a battle, not knowing what to do, issuing a statement — and its leader arriving in time to be imprisoned for an event he had no part in organizing.

The repression was thorough, and its arithmetic is worth holding because it explains the state of the Turin movement when the war ended.

Over 800 militants were arrested and 200 were sent to the front.

Being sent to the front was not imprisonment; it was conscription used as punishment, and in 1917 it carried a real chance of death. So the state removed the militant cadre of the most advanced working-class city in Italy by two routes at once, and did it eighteen months before the strike wave of 1919. When Turin re-emerges in Part II as the centre of the council movement, it is doing so with a leadership that had been arrested, drafted, or newly promoted into gaps left by the missing.

Here the two historians this course leans on hardest disagree flatly, and the disagreement is worth more to you than either verdict.

Start with what they agree on, which is not nothing. Neither treats this as a minor disturbance, and they are the two who disagree about everything else here.

Clark: the Turin riots, in which about fifty people were killed, were the most serious expression of popular discontent to occur during the war; barricades were raised and troops had to be rushed in.

Williams: it was one of the fiercest and bloodiest conflicts in Italian working-class history, and it was consciously internationalist and revolutionary.

Clark says no. The rising was not revolutionary in the sense of being a serious attempt to overthrow the government and seize power, nor even an attempt to end Italy's participation in the war. It arose from a local bread shortage coupled with discontent at long hours in the munitions factories. More sharply still, he places it inside the existing order rather than against it: “The riot was the traditional means of drawing the authorities' attention to grievances.”

Williams says the opposite, and he is not merely asserting a mood — he has the events of card 2 behind him. The workers did not only barricade their own districts: they attacked barracks for arms, and on 24 August they advanced on the centre of the city and nearly broke through, stopping only against machine-guns and tanks. Going for weapons and for the centre is not obviously a way of drawing the authorities' attention to a grievance.

To follow the rest of his case you need one term. The world revolution is the expectation, standard on the revolutionary left in these years, that the Russian revolution was the opening act of a European one — so a rising in Turin counted not as an Italian disturbance but as a second front.

For Williams the rising was consciously internationalist and revolutionary, and Lenin himself described it as part of the development of the world revolution.

Weigh that citation carefully rather than taking it as settling anything. Lenin was in Russia, had never seen Turin, and had a doctrinal interest in counting foreign disturbances into a world revolution. What he establishes is how the rising was read abroad, not what was in the heads of the people on the barricades.

The disagreement is not that one historian has evidence and the other has adjectives. It is that they apply different criteria. Clark asks whether there was a serious attempt to seize power, and finds a bread shortage and a traditional grievance instrument. Williams asks what the action was reaching for, and finds arms, the city centre, an anti-war demonstration and an internationalist temper. Notice that Clark's criterion is the harder one to meet — on it, almost nothing short of a successful coup counts — while Williams's risks counting any angry crowd as revolutionary. Which test you prefer is the real question, and you will need it again in Module 23, where exactly the same choice decides what you make of September 1920.

One caution about this course's own habits before you choose, because the bias is not where you would expect. Almost everything you have been given to think with comes from Williams — functional centrism, the maximum and minimum programme, the drift of "maximalist", the camera del lavoro. Almost every conclusion about capacity comes from Clark — no combat experience, no arms, spontaneity without direction, the riot as a traditional instrument. You are being handed a framework built to show a revolution was refused, and driven towards the conclusion that none was available. Those are compatible, but you should know it is happening.

One piece of evidence bears on Williams's side, and the dates above are what make it work. In the weeks before the rising a Russian delegation toured Italy: Mensheviks and Social Revolutionaries, which is to say the provisional government's own people, and Lenin's rivals rather than his colleagues.

Everywhere the delegation went it was greeted to their visible dismay, with cries of “Viva Lenin!” — and in Turin they got their most enthusiastic reception of all. A little over a week later the city rose.

That is a sharper piece of evidence than a simple pro-Russian enthusiasm would be. Turin crowds were not cheering Russia in general; they were cheering the faction their visitors were losing to, in front of the visitors, two months before that faction took power. It does not prove the rising was made for the world revolution — the cheering crowd and the barricade crowd are not shown to be the same people — but it is real evidence that Williams's internationalist temper was present in the city, and not read back into it afterwards.

Three things the rest of the course runs on.

First, a reputation. Turin became, for both the movement and the government, the city that would rise. That is compatible with the rising drawing no solidarity anywhere else at the time: one is a fact about August 1917, the other about what people afterwards believed Turin was capable of. It is the second that shapes calculations in 1919 and 1920 — including the employers who plan a lock-out there in Module 15, and the leaders who in September 1920 wait to see what Turin will do.

Second, a leader. The arrests took the whole of the section leadership, and into the gap stepped a young Sardinian journalist named Antonio Gramsci — who becomes editor, and by Williams's account the “only reporter”, of the Turin socialist weekly Il Grido del Popolo. He also fought to save the funds of the Turin co-operative alliance from dispersal, and set about rebuilding the movement. Module 8 is his.

The Turin rising precipitated Gramsci into local leadership, and it was after it that his writing acquired new, and marxist, dimensions.

Third, a lesson. The militants who would later found L'Ordine Nuovo, a Turin weekly you meet in Module 8, drew this one: that mass action of this kind was real and enormous, and could not hold a city without an institution able to direct it.

That is nearer Williams than Clark. Clark's claim is that the rising was not reaching for power at all; the ordinovisti conclusion is that it was reaching and lacked the means — which concedes Williams's reading of what the crowd was doing and adds a diagnosis of why it failed. So the premise Part II is built on is not "the rising was only a bread riot" but "the reach was real and the instrument was missing". Hold that, because if you sided with Clark you should carry a doubt about the whole factory council project into Part II, and watch whether it survives contact with 1920.

Part I closes with the movement's best year. In 1919 the Italian workers' movement won more, faster, than at any point in its history: the eight-hour day, membership on a scale no one had planned for, and the largest bloc in parliament. No defeat was inflicted on it all year.

And at the end of it the movement was no closer to power than at the start. That is the puzzle this module sets up and Part II tries to solve — and the solution, when it comes, will be built out of a clause in a metalworkers' contract signed in February.

First, mind the gap behind you. Module 5 left Turin in August 1917 with its cadre in prison or at the front and the war still on. The war ended with the armistice of November 1918, and through 1919 the army demobilised: the arrested came out, the drafted came back, and the men Module 4 described as peasants in uniform went home to a countryside that had been promised land. Almost everything in this module happens because several million men stopped being soldiers at once. The movement Module 5 left in pieces is, sixteen months later, the largest it has ever been.

Begin where the year began, which is not in a negotiating room.

In the early months of 1919 the first great strike wave broke, with clashes with the police and a popular temper Williams calls combative.

No political leader was able to do much at all: it was the CGL and the unions that took the lead, trying to canalize the seething popular discontent into constructive action.

The campaign the CGL launched in January was for an eight-hour day.

Between the beginning of March and the end of April 1919 alone the number of jobless persons rose more than 25 per cent, with the provinces of the Po Valley leading the way.

And prices moved for a reason that had nothing to do with Italy. Through the war the Allies had propped the lira up and held down what Italians paid for what they bought abroad — the allied supports — and in the spring of 1919 they stopped.

The termination of wartime allied supports and price controls had a disastrous impact on the Italian monetary system: when London and New York abolished controls in late March 1919 the value of the lira plummeted, and by June the general price index had quadrupled its prewar levels.

The chain is worth having exactly, because the June risings come out of it: the support ends, the lira falls, and a currency that has fallen buys less abroad. Note that two different measures will appear: this quadrupling is the general index against prewar, while the figure in the June card is Milan's retail food index two months apart. Neither is the other.

Hold that against the victories this module is about to describe. The war industries were cutting back and the demobilised were coming home, so the movement was growing and the labour market was loosening at the same moment. Module 1 gave you Confindustria expecting a recession that would do its bargaining for it; this is that recession beginning, eighteen months early, in the same spring as the eight-hour day.

Membership mushroomed across every federation at once — the CGL, the PSI, the syndicalist USI, the Catholic CIL, the peasant leagues. That is the pressure. The next two cards are what it bought, and what the bill came to.

What the wave won first was the demand Module 2 put on the minimum programme, the list of things worth winning before the revolution. The CGL launched the campaign, and FIOM carried it.

The national metal-workers' agreement of 20 February 1919 established an eight-hour day at last — a 48-hour week, in place of the previous 55, 60 or even 70 hours.

Set that against Module 4's compulsory 70-hour week at Fiat and you have the measure of the victory: the working week cut by nearly a third in three years, with substantial pay rises alongside it. This is what a movement with two million members could extract. It is also the strongest evidence the reformists ever had that accumulation works.

Now read the rest of the contract, because the movement paid for that victory in a currency it did not yet know it would need.

The agreement also recognized the internal commissions — the elected shop-floor grievance committees — and set out in detail what they could and could not do. That half of the contract is Module 7's, because it only means anything once you know what such a body had been able to do before; what matters here is the machinery the same contract built around striking.

If the internal commission and management could not agree, the dispute went to a Commission of Six — three from the employers' federation, three from FIOM.

The procedure in effect ruled out unofficial strikes, since failing to observe it cost a week's pay.

So the price of the eight-hour day was the right to stop work when you chose. A stoppage that had not been through the machinery was now a fineable offence, and the machinery ran through a body half-composed of employers. Module 4 described a wartime factory where striking was a military offence; the peace replaced the court martial with a joint commission and a week's wages.

Note how far that reaches, because it is narrower than it sounds. What was signed on 20 February was the national metal-workers' agreement — a treaty between FIOM and the engineering employers. It binds one trade. It says nothing about a land-workers' league, a building site, a dock, or a crowd outside a bakery.

However, political strikes were exempted from the whole procedure — in accordance with the Pact of Alliance.

Module 3 showed you party and union dividing the world into political and economic action by treaty. Here is that treaty reaching down into a single factory and telling a metalworker which of his grievances he may strike over without penalty. The jurisdictional line was not an abstraction at the top of the movement; it was written into the contract governing the workers who would occupy the plants nineteen months later.

Hold the exemption loosely, though, because it is doing more work than it can carry and this module will come back and take it apart. Everything above assumes that political is a category a worker can act inside. Whether it was depends on who gets to declare a stoppage political — and the answer, when we reach the December strike, is not the one this card implies.

In June and July the action went past strikes altogether — and note that this is a change of kind, not of degree. A bread riot is not an industrial dispute, and no metalworkers' agreement reaches one. Popular insurrections over food prices produced what Williams calls a collapse of authority in some regions, with the movement's own institutions improvising control of towns they had not planned to run. Clark gives the trigger as a sharp rise in prices that spring: the retail food index at Milan went from 287.5 in April to 344.9 in June, on a 1912 base of 100 — the retail measure named two cards ago, and the local face of the currency collapse described there. The rioting began in the syndicalist stronghold of La Spezia — a port on the Ligurian coast — and in anarchist Forlì, the wing of the movement Module 14 takes up.

The rioting spread through other anarchist and syndicalist areas in Liguria and Romagna before reaching the Socialist North and Emilia.

Hold the geography of that spread. It began where the CGL had no members and ended where it had most of them, which is the fact the next two cards have to account for.

So why did it stop? Not because the PSI told it to, though on 5 July the party did warn the populations anxious for immediate solutions against “facile illusions”, and made no connection between immediate solutions and final ones. Williams gives two other answers.

The first is force. The new government of Francesco Saverio Nitti, the liberal prime minister who held office from June 1919, met the risings with military and police repression, and there were several massacres.

The second is the CGL — and this is the one to keep, because the government did not treat force as sufficient.

Nitti's government met the food risings with repression, but placed its faith essentially in the CGL — and, Williams writes, its faith was not misplaced.

The government of the day, facing what Clark calls to all appearances the most serious threat to the established order since the war, did not rely only on troops. It relied on the workers' own confederation — and was right to. The CGL, in Williams's word, duly carried out its “fireman” role and blamed the disturbances on “secessionists”.

Williams adds that the PSI was virtually motionless throughout. Hold that as his verdict rather than as the record: it is a judgement about a party that neither led the risings nor repudiated them, and the last card of this sequence shows the Directorate doing something quite specific while being motionless. Whether that something amounts to motion is the question to carry.

Be precise about that word “secessionists”, because it tells you how the CGL understood what it was doing. The risings had begun in syndicalist and anarchist centres — among the syndicalists who had founded their own federation, the USI, outside the CGL in 1912. So the confederation was not restraining its own members so much as disowning the rival federation it had never contained, and the effect was the same either way: the largest workers' organization in Italy told the country that the risings were somebody else's.

Read that against Module 3. The two million members and the machinery you were shown there are not only a measure of the movement's strength. In June 1919 they were what the government leaned on to see the risings put down, and the government was not wrong to lean.

Hold that sequence next to Module 5. In August 1917 the party could not lead a rising because it did not know one was coming. In June 1919 it knew exactly what was happening and told its people not to be deluded. The two look like the same result — an event outside the institutions, arriving and passing without them.

For 1917 that holds. For 1919 it does not, and the next card is where it comes apart: the disowning was a statement, and a statement cannot stop a rising by people who never took the confederation's orders. Something else was happening while the CGL was talking, and it is the more interesting half of the story.

The CGL's disowning of the risings is not obviously enough to stop anything. If the rioting began among syndicalists and anarchists — people outside the confederation — why would its saying so make any difference at all?

It did not, in the way the question assumes. Nobody was called off. Williams, introducing Spriano's account, describes something else entirely.

The June–July revolt over food prices brought a collapse of authority in the North, and in some areas the CGL was compelled to act virtually as an alternative government.

That is the mechanism the card on June and July left out. The confederation did not talk the rioters down; it was pulled into running the thing they had risen over. The form that took was a set of food committees — local bodies deciding what food there was and what it cost — and in places the rising reached for something larger, declaring soviets, councils of workers' delegates on the Russian model that Module 12 takes properly.

In some towns and villages soviets were set up; almost everywhere shops were sacked, and local food committees were hastily improvised.

Alongside the improvised ones came official food committees, set up by local prefects and sub-prefects — the central government's own appointed officers in each province, the state's man on the spot — to get the situation back under control. So what ended the disorder was not a confederation withdrawing consent. It was the thing the riots were about — who controls food and at what price — passing to committees, some of them thrown up by the rising and some by the state, and in practice staffed by the same people.

Clark describes the committees from above. Corner, writing the history of one province, describes what they did at street level in July 1919 — and both the socialist paper and the prefect agree on the account, which is rare enough to be worth leaning on.

It began with the socialist leaders declining to restrain anybody. They announced that while they would not encourage people to sack the shops, equally — in their own words — “we shall not move a finger, nor say a word, to prevent the starvers of the people from being strung from the lampposts.”

Then agitation against high prices produced hoarding by shopkeepers, and at that point the movement stopped watching and started administering.

Groups of citizens, divided into patrols, forced Ferrara's shopkeepers to sell all their goods at a reduction of 50 per cent, and in front of every shop men with red arm-bands were put on guard to check that the decree of the crowd was respected.

That is the answer to this module's question in miniature. Nobody called anything off. A price was fixed in the street and enforced by the camera del lavoro's own men — the prefect's report calls them vigilante squads from the Camera del Lavoro that began touring the town as soon as the price ceilings were published — until the state conceded the substance of the demand.

Both the socialist paper and the prefect record that the squads disappeared once the prefect gave satisfactory assurances about the maintenance of price ceilings.

Corner adds one sentence about the consequence, and it belongs to Part IV rather than here: a precedent had been created for intimidation. The same streets, the same tactic of armed patrols enforcing a decree on shopkeepers, will be used against this movement within two years, by men who learned in 1919 that it worked.

The movement's local machinery was already staffing those committees. At Turin the local trade union leaders had accepted seats on the official ones, and the shopkeepers had taken their keys to the camera del lavoro rather than to the authorities. The riot was converting itself into administration, with Module 3's institutions administering.

The PSI Directorate answered the food riots by ordering its members — trade unionists in particular — not to sit on the food committees, even at Turin where union leaders had already taken seats and shopkeepers had handed their keys to the Chamber of Labour.

Read that against Module 2 and the shape is exact. Clark says the Directorate could not lead the agitation — it smacked too much of a spontaneous anarchist rising, and the party lacked the organisation and discipline to convert it into a real attempt at revolution — and could not repudiate it either. So it did the third thing: it forbade its own people from the committees where the outcome was being settled. The food riots, Clark concludes, far from beginning the Italian revolution, merely accentuated the differences between the party Directorate and the trade union leaders.

Now settle the question left hanging back at June and July. Williams called the PSI virtually motionless, and on the evidence of these two cards that will not quite hold. In January it was true, and card two said so: no political leader could do much, and the unions took the lead. June is a different case. A party that issues a standing order to its trade unionists, against what those trade unionists were already doing, is not standing still. What is true is narrower and worse for the party — it was moving, and the direction it moved was out.

Williams goes further than that, in the same introduction, and this is his strongest claim about June 1919.

An insurrectionary coup might well have succeeded; instead the PSI and the CGL fought desperately to control the upsurge and direct it towards radical democratic action.

Hold that verdict lightly, because it is the strongest claim any of this course's sources makes about June 1919 and it rests on a counterfactual. But note who is in it. Not the CGL alone, which the last card put at the centre — the party too, and not by inertia. Both institutions worked to bring the rising down to something they could handle, and the question Part II inherits is whether an institution built to do that can ever do anything else.

That is the sharper version of this module's problem, and it is not that the institutions were obeyed. In June the movement's institutions were bypassed by a rising, then quarrelled over whether to administer its results. The next card is the month's other event, and it is the opposite case in every respect.

The international strike of 20-21 July is the opposite case, and the last event of that month: a political action the movement chose, planned and controlled from the start. On 20-21 July an international strike was called across several countries in support of the soviet republics of Russia and Hungary — Hungary had had a communist government since March, and it would fall in August, so this is solidarity with a revolution still in progress — and Clark says some quarters expected it to spark off the revolution outright. Nothing spontaneous about it: in Turin the camera del lavoro convened a meeting on 16 July to settle practical arrangements, and the job of making the stoppage work plant by plant was given to the internal commissions. Keep that detail; it is the first time in this course that the body Part II is about is used as the machinery of a political strike.

The international strike of 20-21 July 1919 had only partial success: the railwaymen defected, and the CGL clamped down on any excess of revolutionary feeling.

The railwaymen appeared once already — Module 1 counted them among the two autonomous unions with no vote at the September 1920 council — and they matter more than that single mention, or their absence from Module 3, suggests.

Clark counts the Railwaymen's Union separately from the CGL's two million: it had 200,000 members of its own and worked closely with the syndicalist USI.

And when Borghi of the USI proposed a National Committee of Revolutionary Action he named five bodies for it: the PSI, the CGL, the USI, the anarchists, and the Railwaymen. So "defected" means this: a union that was one of the movement's five recognised components, and the one that ran the trains a national strike would need, did not come out.

The CGL is the confederation of Module 3, and what Williams says it did is clamp down on any excess of revolutionary feeling. He does not say what that consisted of. But note the position it leaves the CGL in: this was a strike its own side had called, and the confederation spent it policing the enthusiasm of the people who had answered.

Clark is blunter than Williams and calls the strike a failure, one that heightened the impression of maximalist bankruptcy. Williams draws the consequence that matters for Part III: it was this experience that convinced Serrati a revolution was possible only after the organization of a disciplined striking force — a conclusion he will still be acting on in September 1920.

The year ended with the largest victory in the party's history. Module 1 gave you the figures: 1,834,000 votes and 156 deputies in November, the largest bloc in the Chamber. What it did not give you is what the new deputies did with the occasion.

At the opening of parliament in December 1919 the socialist deputies walked out as the monarch entered, and were promptly attacked by crowds of nationalists.

This is maximalism as theatre, in the precise sense Module 2 gave the word: a refusal of the rituals of bourgeois institutions, performed in public, changing nothing about who governed. Note who did the attacking — nationalists, not fascists. In 1919 the fascists are still the fringe of Module 1, and street violence against socialists is not yet their monopoly. The nationalists are a separate and, for now, larger current: the pro-war right, organized before the war and furious after it about the peace settlement, who supplied much of the street muscle that fascism would later absorb. Module 1 has them burning Avanti in April alongside the fascists; here they act on their own.

A general strike followed. Not for a programme or a demand: it was called in response to the assault, a protest stoppage after the fact over an attack on the movement's own representatives.

A general strike in December 1919 should raise an objection. Ten months earlier the metalworkers signed away the unofficial strike: go outside the procedure and it costs a week's pay. How is a stoppage got up at two days' notice not exactly that?

The clause that card told you to keep looks like the answer. Clark's sentence runs: the procedure "in effect ruled out unofficial strikes, for failure to observe the somewhat cumbersome procedure meant… loss of a week's pay; however, 'political' strikes were exempted from these provisions, in accordance with the 'Pact of Alliance'." A protest against an assault on socialist deputies is political by any ordinary reading of the word, so it falls outside the machinery and outside the fine.

Now ask who called it, because the Pact has an answer and the events do not match it. Under the treaty, national economic strikes were to be proclaimed and led by the CGL after hearing the party, while political strikes were to be proclaimed and led by the Party Directorate after hearing the CGL. Constitutionally, then, this one was the party's to order.

The attack on the deputies provoked a spontaneous outburst of strikes and demonstrations by workers in support of the socialists, notably in northern cities such as Mantua.

The spontaneity unsettles the exemption, which is why the question is worth pressing. The exemption is not for strikes that are political in subject matter; it is granted in accordance with the Pact of Alliance, and the Pact defines a political strike as one proclaimed and led by the Party Directorate. A stoppage nobody proclaimed is not obviously a political strike in that sense. It may simply be an unofficial one — which is the category that costs a week's pay.

Did anyone actually lose the week? The sources do not say. Clark gives the rule and the exemption and never returns to what was enforced in December, and no other account here records a penalty. So hold this as a question the course cannot close — but hold Clark's own verdict on the distinction with it: he calls the line between economic and political strikes "clearly an artificial one". The rulebook sorted stoppages by who ordered them, in a year when the largest ones were not ordered by anybody.

Williams names no caller either; he writes only that the deputies were attacked "and there was a general strike". The one action of 1919 that the party was constitutionally entitled to proclaim is also an action that appears to have proclaimed itself.

Which completes the year's pattern, and it is not the one the treaty describes. The strike wave of the early months, the food risings of June and July, and now the December stoppages: three surges out of four, none of them beginning with a decision by the body entitled to take it. The machinery was not overridden in 1919. It was bypassed by events, and then arrived to put its name to them — or, in July, to put them out.

One word in what follows has not been earned. A factory council is, for now, an elected body inside the plant — the institution this module has been circling and Part II is about, grown out of the internal commission and claiming a great deal more than it ever did. That is a placeholder, exactly as Module 3 asked you to treat the word: Module 9 defines it and Module 10 shows one being elected. It is enough to read the claim below, which is the only reason it appears this early.

During the general strike called after the attack on the deputies, Gramsci claimed the councils in Turin mobilized 120,000 workers in an hour, without any preparation.

Hold that figure at arm's length — it is Gramsci's own claim about Gramsci's own institution, made while he was arguing for it, and no independent count is offered. But hold onto it too, because it is the first evidence in this course that the councils could put bodies on the street quickly, and Part II will need to know whether that capacity was real.

Add up what 1919 delivered: the eight-hour day, mass membership, a collapse of authority in whole districts, the largest bloc in parliament. Add them up and the movement should have been close to power. It was not, and Part I has given you the reasons.

The electoral bloc could not govern, because 156 of some 500 seats is not a majority and no one would coalesce with a party pledged to abolish the order it would be governing. The industrial victories were won by a union confederation whose leaders were reformists and whose largest constituent was rural. The insurrections happened where no institution reached. And the party that alone was charged with the maximum programme could issue declarations and nothing else.

And there is a commitment on the record that makes the failure sharper, because the party had signed up to more than it was doing. A Third International was founded in Moscow in March 1919 — the Communist International, or Comintern, created to direct the world's revolutionary parties after the Second International — the body Turati helped found in 1889, back in Module 2 — collapsed in 1914, when its member parties voted for their own governments' war credits — the parliamentary votes that paid for the fighting — instead of opposing the war together. That is the failure the Third International was built to prevent repeating, and it put a choice to every socialist party in Europe. The PSI took it almost at once: meeting in Milan on 18-22 March, the Directorate voted 10 to 3 to adhere. Hold that date against the February agreement: within a month of its own metalworkers' federation signing away the unofficial strike, the party committed itself to the dictatorship of the proletariat.

Whatever its failure to exploit the crisis of 1919, the PSI had adhered to the Third International — which the party itself took to mean creating a cohesive communist party and committing to the dictatorship of the proletariat.

Hold that against everything above. A party that has undertaken to become a cohesive communist party and to bring about the dictatorship of the proletariat, and which then spends the most favourable year in its history issuing declarations, is not merely ineffective — it is in breach of its own undertaking. Williams notes that one man understood what adherence implied from the first, and his name is Amadeo Bordiga; you meet him in Module 13.

That is the state of the movement entering 1920: enormously strong in every dimension that can be measured, and structurally unable to convert any of it into a decision. Which is exactly the gap a new institution might fill — one that sat where the workers actually were, in the factory, and could act without waiting for a treaty between a party and a union to tell it whose grievance this was.

The internal commission already existed — not in embryo, and not new. It was the body the last card was reaching for, its history reached back well before the war, and the February agreement did not create it but recognized it and hedged it about with conditions. Module 7 goes back and finds where it came from, which turns out to matter: a body with a history is a body with prior claims.

Part I ended with a movement at maximum strength and minimum capacity: strong everywhere it could be counted, with no body able to convert that strength into a decision. Part II is about the one institution that looked like an answer.

It did not arrive as a revolutionary invention. It arrived as a grievance committee, set up by agreement with employers, to argue about punishments and dismissals. How that became a claim to run the factory — and then, on the claim of the men who theorised it, a nucleus of a future workers' state — a state whose governing bodies would be these factory institutions rather than a parliament — is the story of this part.

Part II covers roughly a year, from the spring of 1919 to April 1920, and almost all of it happens in one city. Modules 8 and 9 supply the theory and the definition, Modules 10 to 12 the actual building of the councils in Turin, Modules 13 and 14 the two serious critiques from the left, and Modules 15 and 16 the counter-attack that ends the part.

You will have noticed that this overlaps Part I, and it is worth being told why rather than left to wonder. The two parts are cut on different principles. Part I was chronological and national: it ran to the end of 1919 and asked what the whole movement did with the year. Part II is neither — it follows one institution, in one city, from before the war to its defeat. So 1919 is covered twice on purpose, once as the movement's year and once as the council's. Module 6's December strike will come back in Module 12 looking quite different. Where the two accounts of the same month disagree, that is the course showing you its working, not losing track.

This module is the prehistory: where the raw material came from, and why it was already contested before anybody wrote a word about councils. It runs from 1906 to the end of the war, before Part II's own clock starts.

The internal commissions began early in the century as temporary, ad hoc committees to put forward complaints concerning punishments, dismissals, and the like.

The big questions of wages and hours stayed with the unions. The commission handled the things that happen to an individual worker on a particular morning: a fine, a sacking, a dispute about how a rule was applied. It was elected by the union members on the shop floor, and it existed to make an agreement work rather than to challenge it.

Everything the council movement later claims is built on a body whose founding purpose was to keep a contract running smoothly — which is exactly why the employers agreed to have one.

The first time a permanent commission was recognized, it was given powers wider than any it would hold again.

In October 1906 the first permanent internal commission was recognized in an agreement between FIOM and the Turin car firm Itala, whose Article 19 provided for five workers holding office for the agreement's three years.

Three features of that deal matter, and the first is in Article 19 itself. Five workers, holding office for the agreement's three years — that is what the word permanent is doing in this card. The commissions of the last card were convened for a grievance and dissolved with it; a body with a fixed size and a fixed term outlives the dispute that produced it, which means it can accumulate practice, precedent and standing. An institution begins where the tenure clause is.

Second, a closed shop was set up alongside it — that is, employment in the plant was restricted to union members, which gave the union real control over the workforce. And third, on procedure the agreement was extraordinary: all conflicts arising from the contract, including wages and hours, were to be discussed first between the internal commission and the management, with the union called in only if they failed to agree.

Now set 1906 against what the February 1919 agreement gave the same institution, because Module 6 deliberately left those clauses here. In 1919 the commission was recognized and given real standing: it could negotiate on collective as well as individual grievances, it held office for a year, and it got an office of its own inside the factory. Then the contract added a condition — it was to fulfil its tasks only outside working hours.

Be precise about what changed between the two, because it is not the order. In both agreements the shop-floor body goes first and someone else hears the appeal. Three things differ. The scope shrinks: in 1906 the commission took all conflicts including wages and hours, in 1919 it takes grievances while wages and hours belong to the union. The tenure shortens, from Article 19's three years to one. And a new limit appears that 1906 had no equivalent for: the commission may work only outside working hours.

So what 1919 grants in recognition it takes back in reach. That is the shape of the quarrel Part II is about, and it is worth noticing that on this evidence the high-water mark of shop-floor authority is 1906, not 1920.

And the other side moved in the same year, which is the part usually left out.

In the same year the whole working class of the city struck in support of the textile workers and, in response, the employers formed a Turin Industrial League.

So 1906 is not simply a workers' high-water mark. It is the year both sides built something permanent, and the employers' body outlived the workers'.

The Itala agreement did not last, ended by the bitter industrial struggles of 1907-8, after which the commissions became temporary bodies again — with very limited functions and well under the control of the Unions.

The struggles of 1907-8 have been a cause in this module without a content. The year after Itala the organized employers struck back: they extended the League's cartel — their agreement to act as one employer rather than compete for labour or settle separately — from the city to the province, tried to co-ordinate with Genoa, Milan and Terni, and ran — in Williams's phrase — into headlong collision with syndicalist-inspired resistance. The workers lost. A series of shattering defeats coincided with the onset of a depression that rationalized the car industry, meaning fewer firms and fewer men in them.

And then the detail that closes a circle. Williams: "On the ruins of working-class militancy, the first Confindustria, general confederation of industrialists, formed at Turin on the model of the League." Module 3 dated Confindustria's first formation to 1910 and told you it copied the Turin Industrial League; here is what it was built on. The employers' national body was the victory monument of 1907-8 — and it is the same body that, in Module 15, will take the councils apart.

The first Confindustria formed at Turin in 1910, on the model of the Industrial League and — in Williams's phrase — on the ruins of working-class militancy after the defeats of 1907-8.

So the commissions did not simply decay. They were beaten down in a defeat that also produced the organization which beat them again thirteen years later, and the shop-floor body spent the intervening years as the unions' creature.

By 1913 the commissions had been pushed further down still: agreements in the Turin automobile industry set up separate arbitration bodies inside the factories containing representatives of both employers and workers, and the commissions tended to be subordinated to these or merged into them. A purely workers' body on the shop floor had effectively ceased to exist. That was the position when the war began.

Then Module 4's regime changed the problem. Under Industrial Mobilisation the plants filled with new workers under quasi-military discipline, strikes became a military offence for the third of them who were military or exempted, and union officials were distant from a shop floor that now generated disciplinary grievances daily.

The sweeping changes brought by Industrial Mobilisation revived the need for some kind of purely workers' body to represent grievances — especially disciplinary ones — to management.

The employers wanted it too. Wartime expansion meant they were more likely to concede on minor matters so that production could continue, and Clark states the consequence plainly: the existence of a workers' body within the factory was an advantage to management as well as to the workers themselves.

Clark does not say the commission was a management device; he says it suited both parties at once. An institution that genuinely serves opposed interests is not a trick played by one side on the other — it is a piece of ground they both want to stand on, and the question is who ends up occupying it when their interests stop coinciding.

Very shortly after Italy entered the war the institution was virtually ubiquitous in the metal-working factories of Lombardy.

That was Lombardy — Milan and its industrial belt — and this course is about Turin, so take the Turin evidence separately. Clark's Turin case is the Diatto Bros. works in 1917, where at the end of a dispute the workers' commission "was given a mandate to stay in office so as to check on complaints received and ensure that the agreement obtained is observed." Read the grammar before the meaning: Clark's sentence is passive and names nobody as the grantor. The commission was given a mandate — by the workers who elected it, or by the settlement that ended the dispute, and the quotation does not say which.

That matters more than it looks, because the two readings point opposite ways. If the workers kept the body in being, this is the 1906 tenure principle reappearing in the middle of a war — a shop-floor body outlasting the grievance that created it, which is the whole argument of this module. If the settlement conferred it, it is an employer's convenience: someone to check compliance, kept on for the employer's benefit as much as the workers'. This course cannot tell you which, and it is one firm in one year. Take it as consistent with the argument rather than as evidence for it.

The pressure of wartime disputes broke down the distinction between ad hoc committees appointed for specific purposes and permanent workers' representatives inside the factory.

Emilio Colombino and Mario Guarnieri, heading the Turin metalworkers, were stressing the importance of these bodies by early 1917: most conflicts, they wrote, were settled without any decree from the Regional Committee of Industrial Mobilisation, because the majority of cases brought before the Committee ended up handled by the same workers' commissions with a Committee delegate merely present. "The Committee's ordinances were few."

Hold that beside Module 4. The Regional Committee was the state's machinery for war industry — an officer in the chair, employers and workers in equal numbers — and in Turin the real work of settling disputes was being done instead by shop-floor bodies that machinery had not created, with the official organ reduced to attendance.

And note what the two wartime and postwar bodies have in common, which is less than a shape and more than a coincidence. The Regional Committee of Module 4 sat an officer and two economic experts alongside two men from each side; the Commission of Six of Module 6 was three and three with nobody named as chair. They are not the same design. What they share is the direction of travel: in both, a dispute leaves the floor it arose on and goes up to a joint body to be settled. The council movement of Part II is, among other things, an attempt to stop sending the question upstairs.

Put the three conditions together and you can see what the commission became without anybody deciding it should. It was inside the plant, where the union's officials were not. It was elected — by the union members among the workforce, which in a closed shop means everyone and elsewhere does not, a distinction Module 10 turns into the whole quarrel — and it therefore contained no employers, unlike the joint arbitration boards of 1913. And it was the only channel available at a moment when the strike had become a military offence for the third of the auxiliary workforce who held military or exempted status — and an offence against factory discipline, backed by dismissal, for everybody else.

An institution that is the only one present when a grievance arises does not stay within its remit for long. FIOM's own journal spent the war reporting successful negotiations, usually on disciplinary matters, carried through by internal commissions. Each of those is a small precedent for a body arguing with management on its own account, and by 1918 there were years of them.

The rank and file means the ordinary membership as against the officials — a distinction that is about to matter more than any other in this course.

In November 1918 the FIOM congress at Rome was disrupted by a revolt of the rank and file focusing on the internal commissions.

That is the moment the institutional question becomes a political one — and it is worth taking apart, because what the revolt actually demanded is not what the sentence above suggests.

The Rome revolt began with a government committee. In 1918 the government of Vittorio Emanuele Orlando — the wartime prime minister who took office after Caporetto and led Italy to the peace conference — set up the Commission for the Study of Measures for the Post-war Period — universally known as the Commissionissima, roughly "the great big commission" — to draft proposals for the economy and society the war would leave behind. Much of the CGL and the parliamentary group wanted to take part in it.

Whether to sit on it split the movement, and the split ran along Module 2's line exactly: to reformists it was a chance to shape the postwar settlement, to the party Directorate it was collaboration with a bourgeois government. Two things came out of that quarrel and this course needs both. Rigola resigned the leadership of the CGL over it — and his successor D'Aragona's first task was to negotiate the Pact of Alliance of Module 3. The other thing happened at Rome.

Bruno Buozzi, FIOM's secretary, took part in the Commissionissima's work. A meeting of Turin metalworkers in August condemned him for acting without consulting them, and his reply is the sentence that made the congress a revolt: unions and their secretaries, he said, "even if they do not seem to be very militant or class-conscious, are always far more so than the anonymous masses".

The Turin delegates to the Rome congress answered with a manifesto: the leaders should follow the tactics and the paths laid down by, and wanted by, the masses — "otherwise they should quit their posts in favour of men better able to interpret the needs and wishes of the masses." Their programme demanded greater participation by workers in deciding union policy, greater combativity, and a revision of the role of the internal commissions.

Williams's phrasing is worth keeping: the anonymous masses replied in November. So the demand was not, in the first place, for power against the employers. It was for a union that did what its members wanted, with reformed commissions as the instrument — greater say through the shop floor, running parallel with a rejection of union collaboration with bourgeois society. The commissions became the vehicle for a quarrel about who the union belonged to.

What makes Rome genuinely instructive is that the leadership was not defending the status quo. It was proposing to expand the commissions too.

Union leaders had spent the summer of 1918 aware of the growing cleavage between their members and themselves, and had begun moving to re-integrate the commissions into union structure. At Rome the FIOM spokesman Colombino backed proposals to turn them into permanent organs of workers' control, meaning surveillance over wages and discipline, under the supervision of the unions.

That is the same Colombino who, three cards ago, was heading the Turin metalworkers and testifying that the commissions were settling most disputes on their own. He is on the leadership's side of this quarrel, and the delegates revolting against the leadership were Turin men too. Do not read the congress as Turin against the rest: the split ran through the Turin delegation itself.

At Rome the FIOM spokesman Colombino backed turning the internal commissions into permanent organs of workers' control — surveillance over wages and discipline — under the supervision of the unions.

And he argued for it in frankly reformist terms: an improvement in capital-labour relations, and "the elevation of workers from automata to citizens". He praised the British Whitley councils — the joint employer-worker committees set up in Britain from 1917 to formalise consultation in industry. Williams calls this a classic statement of the CGL case, the same case behind its demand that a Constituent Assembly — an elected body convened to write a new constitution, which the reformists were already pressing for — be built instead as a "labour parliament", and the same case that will shape its response to the occupation of the factories in 1920.

So set the two proposals side by side, because the shape of Part II is already here. Nobody at Rome wanted the commissions left as they were. The militants wanted them revised so the shop floor could bind the union; the leadership wanted them made permanent so the union could supervise the shop floor. The quarrel was never whether to strengthen the institution. It was who would end up holding it — which is the question the next card poses, and the question Part II spends a year fighting over.

Everything in Part II follows from an ambiguity that the last three cards have built up, and that nobody in 1919 could resolve by argument alone.

On one reading the internal commission is an organ of workers' power. At Itala in 1906 it took every conflict in the plant, wages and hours included, before the union was called in. Through the war it accumulated years of successful negotiations with management on its own account, reported in FIOM's own journal. And in November 1918 its constituents used it as a weapon against their own union leadership at FIOM's congress — which is not how people treat a body they regard as the management's.

On the other reading it is a management convenience: created by agreement with employers, originally confined to punishments and dismissals while the union kept wages and hours, valued because it kept production running, and — after February 1919 — restricted to hours when no production happens and made to escalate through a body half composed of employers, under a contract that also imported the Pact of Alliance to tell a worker which grievances he could strike over. Module 6's Rigola reading sharpens this: the employers conceded hours to buy a free hand over the job itself.

Both readings have real evidence, and Clark's sentence says why: the thing was useful to both sides. So the honest form of the question is not "which was it really" — origin does not fix function, and an institution can change hands. It is: when the interests of production and of the workers came apart, which way did the commission face? That is answerable by evidence, and Part III answers it.

What happens next, then, is not a discovery but a fight over an institution both sides had reason to want. Module 8 introduces the group that took the first reading and made a programme of it, and Module 9 gives the two changes — who votes, and what the body is for — that turn an internal commission into a factory council.

One loose end before Part II proper, and it is Module 6's. That module ended by promising an institution that could act "without waiting for a treaty between a party and a union to tell it whose grievance this was". On the evidence of this module, it could not: the February contract wrote the Pact of Alliance directly into the commission's procedure. So the council movement does not inherit a body that has escaped the machinery of Part I. It inherits one the machinery has already caught, and its programme will have to be a way of breaking out.

Module 7 left the internal commission as an institution both sides had reason to want, and asked which way it would face when their interests came apart. That question is answered by events, in Part III. This module is something different: the first serious attempt to decide the answer in advance — a programme, written by four young socialists in Turin, for taking the commission and making it the germ of a workers' state.

Keep the distinction. What follows is advocacy, not evidence. It tells you what one group wanted the institution to become; it is not a finding about what it was.

It is worth saying at the outset that they did not begin with a plan. They began with a magazine, and by their own account they were not sure what it was for.

You met the central figure in Module 5: Antonio Gramsci, the young Sardinian pushed into the leadership of the Turin section when the August 1917 arrests removed everyone above him. Around him were Angelo Tasca — who had founded the Turin socialist youth section back in 1909 and had brought Gramsci into the party — along with Palmiro Togliatti and Umberto Terracini.

The review was initially financed by 6,000 lire provided by Angelo Tasca, the only member of the group with contacts good enough in the local labour movement to raise any money.

The first issue of L'Ordine Nuovo — "The New Order" — appeared on 1 May 1919, described as an organ of proletarian culture.

Hold onto that description, because it is not the description of a body preparing to take over the factories. Gramsci himself was blunt about the founding, writing a year later:

“When, in April 1919, we decided, three, four or five of us, to begin publishing this review, the Ordine Nuovo… who were we? What did we represent? What message did we bring? Alas! The only thing that united us in our meetings was a vague passion for a vague proletarian culture.”

That is a useful corrective to any account in which the council movement descends fully formed from a theory. The theory arrived seven weeks after the magazine did, and it arrived because the group went looking for something the magazine could actually be about.

On 21 June 1919 the review published an article called "Workers' Democracy", and it is the founding document of everything in Part II. Its first move is the one to understand, because everything else follows from it.

Gramsci's opening claim: the socialist State already exists potentially in the institutions of social life characteristic of the exploited working class.

It inverts the ordinary picture. On the usual account a revolution seizes the state and then builds socialist institutions. Gramsci says the institutions are already here, in the ordinary organizations working people have built for themselves — the kind Module 3 catalogued, the chambers of labour, the co-operatives, the leagues, and above all the workshop with its internal commission — and the task is to link and order them into a hierarchy of competences until they constitute, in his phrase, a genuine workers' democracy here and now, standing in active opposition to the bourgeois State and prepared to replace it in its essential functions.

So revolution stops being a date in the future and becomes a construction project in the present. That is the reading this course will follow — but weigh the word potentially before you accept it. Any institution can be said to contain anything potentially, and Gramsci offers no test for when the potential becomes actual. A hostile reader would say a premise that cannot fail is a premise that can support any conclusion, and that objection is not answered anywhere in the article.

Treat him, in other words, exactly as this course has treated Williams, Clark, Lenin and Rigola. He is a participant with a programme to advance and a faction to build, writing seven weeks after founding a magazine he has just told you had no clear purpose. That is not a reason to discount him. It is a reason to read him as an author rather than an authority — which, given that Williams wrote the history of Gramsci's own movement and supplied most of this course's vocabulary, matters more here than anywhere else in the course.

Before naming his institution, Gramsci explains why the two organizations Part I described are not it. His objection is not that they are treacherous. It is structural, and it should sound familiar.

The social power of the Party and the Confederation, Gramsci wrote, reaches the great mass of workers only indirectly — by prestige and enthusiasm, authoritarian pressure and even inertia.

That is Module 2's functional centrism and Module 3's wiring diagram, stated in 1919 by a participant rather than in 1975 by a historian. And Gramsci draws the conclusion that the machines cannot be reformed into the answer: it would take the Socialist Party and the trade unions years, even decades, to absorb the whole working class, and these two institutions would not be identified with the proletarian State in any case.

He does not propose abolishing the party. Its role remains real but changed — the organ of communist education, the depository of doctrine, the force harmonizing the disciplined energies of the workers and peasantry. What he denies the party is the claim to be the coming state. The state has to be built out of institutions the whole class actually inhabits, and the party is not one of those.

Now the specific proposal, and it lands exactly on Module 7's institution.

The internal commissions, Gramsci wrote, are organs of workers' democracy which must be freed from the limitations imposed on them by the entrepreneurs, and infused with new life and energy.

And then the sentence that turns a grievance committee into a revolutionary programme: “Today the internal commissions limit the power of the capitalist in the factory and perform functions of arbitration and discipline. Tomorrow, developed and enriched, they must be the organs of proletarian power, replacing the capitalist in all his useful functions of management and administration.”

Set that against Module 7's two readings, and read the whole sentence rather than half of it. Gramsci gives the commission both descriptions at once: today it "limits the power of the capitalist" — the workers'-power reading — and performs "arbitration and discipline" — the management-convenience reading. He is not choosing between them. He is saying the institution is currently doing both jobs, and that the second can be made to give way to the first.

Now notice the grammar, because it is where the argument is thinnest. Every operative verb in the proposal is an imperative: the commissions must be freed, they must be the organs of proletarian power, the workers should proceed at once to elections. Gramsci says the limits were imposed by the employers; he does not, in this article, say by what means a body whose remit is fixed in a signed contract acquires a different remit. That gap between must and can is the whole practical question of Part II, and the founding document does not address it.

One more thing a careful reader should refuse to let pass. Gramsci attributes the commissions' limitations to "the entrepreneurs" — but Module 6 showed you where the sharpest of those limitations came from: the February 1919 agreement that FIOM signed, in exchange for the eight-hour day, and which imported the Pact of Alliance into the shop floor. Blaming the employers alone is convenient for a proposal that needs the workers' own institutions to be blameless. It is the first sign that this is a partisan document as well as an analytical one.

The article does not stop at principle; it specifies a machine. Delegates were to be elected in the workshop, by category — manual workers, clerical workers, technicians — building upward through a series of elections to a committee representing every aspect of the work.

The aim, Gramsci proposed, should be to elect one delegate for every fifteen workers, divided into categories as is done in English factories.

That parenthesis matters. The ordinovisti — as the group around the review came to be known — were consciously borrowing from the British shop-stewards' movement, the wartime shop-floor delegate organization in British engineering plants, which the review described at length that July. Their intellectual honesty about the borrowing is worth noting, and so is the awkward fact they do not discuss: Britain produced a vigorous shop-stewards' movement and no workers' state whatever. Borrowing a mechanism from a country where it did not produce the intended result is not, on its face, an argument for the mechanism.

Above the workshop, ward committees were to co-ordinate all the proletarian energies of a district, taking in waiters, cab-drivers, tramwaymen, railwaymen, road-sweepers and clerks — an expression of the whole working class living in the ward, with the power to order the immediate cessation of all work throughout it. The slogans were "All power in the workshop to the workshop committees" and "All State power to the Workers' and Peasants' Councils".

Step back and see the shape of it against Part I. The problem was a movement with enormous strength and no body able to decide. Gramsci's answer is to build a new body, in the one place where the whole class is already assembled and already has an elected institution — the factory.

It is an ingenious solution, and it has four vulnerabilities. It depends on an institution the employers agreed to and the February contract had just restricted. It rejects the union's claim to lead without saying what happens when the union objects — and there are two unions in that sentence, which the programme has to get past separately: FIOM owns the internal commissions and signed the February agreement, while the CGL owns economic action by treaty and has two million members behind it. Third, the peasants: Gramsci's slogan is "All State power to the Workers' and Peasants' Councils", and the party is to harmonise the energies of workers and peasantry, but the mechanism he specifies is entirely industrial — a delegate ratio for a workshop, categories for manual workers, clerks and technicians, ward committees for a city district. There is no rural machinery in it at all, and Module 3 put 900,000 land-workers against 160,000 metalworkers.

The fourth is the one the article never mentions, and it is the largest. Nothing in "Workers' Democracy" concerns the army, the police or the prefects. Modules 4 and 5 established that the industrial working class had no combat experience and had not armed itself, that demobilisation would produce an army largely of peasants hostile to exempted industrial workers, and that in August 1917 peasants in uniform shot Turin workers in the street. A theory in which the replacement state arrives when it becomes more real than the old one has quietly declined to name the moment at which somebody starts shooting.

Module 9 takes the proposal apart properly: what exactly distinguishes a factory council from the internal commission it grows out of, and why the two changes involved — who votes, and what the body is for — turn a contractual grievance committee into a claim on sovereignty.

The course has been deferring this term since Module 3, where you were told to treat "factory council" as a placeholder. Here is the definition, and it is worth the wait only if you can say precisely what it excludes.

A factory council is a body of delegates elected inside a workplace by the people who work there, organized by unit of production rather than by trade or union membership, claiming authority over production itself rather than over the terms on which labour is sold. Two things in that sentence do the work: who belongs, and what it is for. Everything else is detail.

Module 7 left the internal commission elected by the union members on the shop floor, often in practice chosen by union officials with the election a formality. The council movement's founding document changed the electorate, and it is worth having the mechanism exactly.

The Programme of the Workshop Commissars called for the internal commission to be elected by workshop commissars, who were themselves elected by universal suffrage of workers at their workplace, by workshops.

So the answer to a question the last two modules have left open — what becomes of the internal commission once a council exists — is that they are the same body transformed. Williams calls the result revolutionized internal commissions, and says these were the leading organizations of the council movement. The council is not a rival institution built beside the commission; it is the commission with a new electorate underneath it and a new claim on top.

That change of electorate sounds procedural and is a change of principle. Gramsci states it as a contrast between two kinds of institution:

The Councils are historical products, brought into being by the need to master the apparatus of production; the trade unions and the Party are voluntary associations.

You join a union. You are in a council, by virtue of working where you work. The union's boundary is a decision each worker makes; the council's boundary is drawn by the factory gate. That is why the council can claim to represent everyone in the plant while the union represents only its members — and it is why the council movement will collide with the unions rather than complement them.

It also solves a problem Part I left open. Module 3 gave you a movement split across rival federations — socialist, Catholic, syndicalist — each with its own paper and its own jurisdiction. Inside one plant those divisions are invisible to the machinery: a council elected by everyone at the bench contains Catholics, syndicalists and the unorganized alike. Gramsci wrote that the Council realizes in practice the unity of the working class, and this is the mechanism he meant.

But notice at once what the factory gate excludes as well as includes, because the course has already given you the number. Federterra's 900,000 land-workers do not work in plants and cannot be enrolled by workshop. An institution whose electorate is defined by a factory can unify everyone inside the factory and, by the same stroke, cannot reach the largest single body in the movement. The unity the council achieves is real and it is bounded, and Gramsci's language — the unity of the working class, a congress open to the whole Italian proletariat — is wider than his mechanism.

The deeper change is in purpose, and Gramsci puts it as a distinction between two ways of describing the same man.

The proletarian dictatorship can only be embodied in an organization specific to the activity of producers, not wage-earners, the slaves of capital — and the factory Council is the nucleus of that organization.

A wage-earner is someone who sells labour by the hour, and the institution proper to him is the union, which exists to raise the price. A producer is someone who makes things, and the institution proper to him is a body that runs the making. The same person is both; the question is which description his organization is built on.

Gramsci's argument for preferring the second is that it is the one that outlasts capitalism. Be exact about the claim: what he calls permanent is the council's reason for existing, which lies in the labour process, not the survival of any particular council. The union's lies in wages and class divisions, which are transitory and, as he puts it, the very thing we are trying to supersede. A union that won everything it wanted would have negotiated the best possible price for labour power and left the wage system intact. A council that succeeded would have abolished the buyer.

Hence the claim the whole movement rests on: the Factory Council is the model of the proletarian State.

And a consequence Gramsci draws with some relish: in the council as in the workers' state, the concept of citizen gives way to the concept of comrade. That is not decoration. A citizen is an individual bearing rights against a state; a comrade is a member of a body engaged in a common task. The council is meant to be a school that teaches people to be the second thing.

The machinery is simpler than the theory. Workers in each workshop elected a workshop commissar — a delegate for their own crew or craft — and the commissars of a plant together formed its council.

Note that the council is not a mass meeting; it is a delegate body built from the shop floor upward, one crew at a time. Gramsci was explicit that this preserves the craft rather than dissolving it: while the trade union is based on the individual, he wrote, the Council is based on the concrete, organic unity of the craft as it is realized in the discipline of the industrial process.

That is a real difference from the union's way of counting. A union federates individuals who happen to share a trade across many employers. A council assembles the actual working units of one plant in the proportions in which they contribute to making the thing. It is organized, in other words, by the shape of the work.

All sectors of the labour process are represented in the Council in proportion to the contribution each craft makes to the manufacture of the object the factory is producing.

The point of the structure is educational as much as institutional, and this is the part most easily missed.

Once the Councils exist, Gramsci wrote, they give workers direct responsibility for production, instil a conscious and voluntary discipline, and create a producer's mentality — the mentality of a creator of history.

Set that beside Module 8's claim that the socialist state already exists potentially. The council is how the potential is meant to become actual: not by decree but by people learning, in the course of running a workshop, to behave like the governing class they are supposed to become. The proposed national congress of councils was to be open to the whole Italian proletariat, represented by expressly elected delegates and not by union officials — the distinction enforced right up to the top.

Three objections, in ascending order of seriousness. Hold them through the rest of Part II, because Modules 13 and 14 will bring critics who make versions of the last two.

First, the change of electorate does least where the movement is strongest. Module 7 noted that under a closed shop — employment restricted to union members — "elected by the union members" and "elected by the workforce" describe the same electorate. So in the best-organized plants the franchise reform is null, and the change of principle is available in inverse proportion to union strength. That is an awkward property for a reform meant to unify the class.

Second, and worse: the definition's third clause is a claim, and nobody has shown how the claim becomes an authority. A council is defined as claiming authority over production. But the body it is built from has a remit fixed in a signed contract — the February 1919 agreement, which confined it to hours outside production and routed its disputes through a panel half composed of employers. Module 8 showed that Gramsci writes in imperatives: the commissions must be freed. Neither he nor this module has said by what mechanism a body acquires a remit its contract denies it. Until that is answered, "factory council" names an intention rather than an institution.

Third, the strongest: the definition may not pick out anything new. Module 7 recorded that at Itala in 1906 the shop-floor body took all conflicts arising from the contract, wages and hours included, before the union was called in — and concluded that the high-water mark of shop-floor authority was 1906, not 1920. If the peak of actual shop-floor power predates both of the changes this module calls decisive, then the changes are not what produces shop-floor power, and what distinguishes a council from a commission is its ambition rather than its capacity.

The three objections above all attack the franchise or the claim. This one attacks the purpose. If the council organizes the worker as producer, its natural business is output, discipline and the organization of work — which is management's business. Gramsci embraces this: councils are to give workers direct responsibility for production and instil a conscious and voluntary discipline. Set that beside Rigola's suspicion from Module 6 and the question is obvious: what distinguishes a producer's mentality from a well-run productivity scheme?

The reply available to the ordinovisti — though I have not found them making it in these words — is that the difference lies in who holds the authority, not in what the work looks like. Note that this reply concedes the second objection above: it is about authority, and authority is exactly what has not been shown to transfer.

So carry three things forward. The two changes are what make a council a claim on sovereignty rather than an improvement in industrial relations. The claim was contested from the moment it was made, and Module 10 shows the first fight, which is about the franchise. And the objections above are not answered in this module because they cannot be answered by argument — they are questions about what a council could actually do, and Part III is where councils meet employers who have stopped co-operating.

Module 9 claimed that changing the electorate from the union members to the whole workforce was a change of principle rather than of procedure. In August 1919 that claim was tested twice, in two plants of the same company, eight days apart, under two different franchises.

It is often told as though it were a controlled experiment. It is not one — nobody assigned the franchises, the participants chose them — and one of the two outcomes was never recorded. This module gives you the episode and then takes the evidence apart, because the Fiat-Centro/Brevetti pair is the most-cited support for the council idea and it will not carry the weight usually put on it.

The first move came from below and was not the ordinovisti's doing. At the main Fiat works the existing internal commission resigned, after repeated attempts — Clark does not say by whom — to force it out on the grounds of its old-fashioned ideas and its close connection with FIOM; it was alleged that some of its members even held union jobs. A temporary commission took over with a mandate to organize the election of a commissar in each work unit, and the vote came on 23 August.

The work unit was the reparto — the shop or department — and at Fiat-Centro there were 42 of them.

Clark is careful to say this was not a simple victory for L'Ordine Nuovo's ideas. The temporary commission wanted commissars mainly to keep local union officials out of the plant's affairs — a turf motive, not a revolutionary one. And when it came to the vote, it kept the old rule.

At Fiat-Centro the election of the commissars, in which only Union members were allowed to vote, failed to produce an anti-Union majority.

The new internal commission the commissars then chose defined its own job narrowly: assisting in controversies arising in the workshops over agreements between the Industrial League — the Turin employers' association Module 3 named as the model for Confindustria — and FIOM, in which, it added, it had confidence. Clark records the consequence: the commission at the main Fiat works proved the most conservative in Turin over the following months.

On 31 August, at the Brevetti-Fiat plant across the city, Gramsci addressed a meeting and elections for commissars were held in each work-unit on the other rule. The rule is stated explicitly in Williams's chronology, which matters — it is not being inferred from how many people showed up.

At Brevetti-Fiat the workshop commissars were elected with all workers having the right to vote, and union members only having the right of candidature.

Williams's rule says "union members"; Clark, describing the same election, is more specific — only members of FIOM were eligible. In a plant that could also contain Catholic or syndicalist unionists, the difference matters, and the course follows Clark's narrower version because he is the closer account.

The Brevetti arrangement was hailed by L'Ordine Nuovo as a triumph of its own principles.

The turnout is worth having as well: Clark records that all but three or four of the plant's 2,000 workers voted. That is a real fact about enthusiasm — a plant of two thousand in which essentially everybody turned out. But be careful what it can prove. Turnout measures participation, not eligibility, and it certainly does not tell you how the two thousand voted.

Module 9 gave you the council principle in its pure form: the electorate is the workforce. Brevetti-Fiat delivered half of it. Everyone could vote; only FIOM members could stand.

Think about what each half does. Opening the vote to everyone makes the body representative of the plant — the unorganized, the Catholics, the syndicalists all count. Restricting candidature to FIOM members means that whoever is elected is, necessarily, a member of the socialist metalworkers' union. The electorate is the class; the eligible are the organized.

So the arrangement L'Ordine Nuovo hailed as a triumph of its principles was a compromise in which the union kept a monopoly of candidature. You can read that two ways, and both are defensible. Generously: a transitional device, widening the base while the organized minority still supplies the cadre. Ungenerously: the council movement conceding, at its first practical success, precisely the point Module 9 said was a change of principle — and conceding it to the institution it claimed to be superseding.

Now put the two plants side by side, because the contrast is the module's whole argument.

Same firm. Eight days apart. Both elected commissars by work unit, so the structure was identical, and the electoral rules differed. Fiat-Centro, on the union-members franchise, returned a commission that declared its confidence in the existing agreements and became the most conservative in the city. Brevetti, on the wider franchise, is described by Clark only as the plant where the L'Ordine Nuovo campaign had rather more success.

Now stop and notice how thin that is, because the comparison is famous and the evidence for it is weaker than its reputation. The two outcomes are not measured on the same thing. For Fiat-Centro we are told the political composition and the subsequent conduct of the body elected. For Brevetti we are told the turnout, the rule, and a historian's summary judgement — no vote breakdown, no composition, nothing about what the resulting commission did. You cannot properly compare an outcome with a turnout, and this course should not pretend otherwise.

There is a second and sharper problem, and it points the causal arrow the other way. At Fiat-Centro the election was organized by a temporary commission whose motive, on Clark's account, was to keep local union officials out of the plant — and that officials-averse body chose to keep the union-members franchise. (Hostility to the local officials, note, is not the same disposition as hostility to the union: the commission this election produced declared its confidence in FIOM's agreements.) If the participants had believed the franchise decided outcomes, that choice makes no sense. The likelier reading is that in each plant the rule was set by whoever already held the balance of forces there, in which case the franchise reflects the local balance rather than producing it, and the correlation is largely an artefact.

So take the pair as suggestive and not as proof. It shows two rules in use, and the movement treating the wider one as a victory. It does not establish that the rule caused the result, and Module 9's claim — that the change of electorate is a change of principle — was never the kind of claim two election results could settle anyway.

And the fight was not chiefly with the employers. Both elections were internal to the workers' side. The question at issue — may a man who has not joined the union help choose who speaks for his shop? — is a question about the union's standing, and it is the question Module 11 shows becoming an open war.

Whatever the compromises, the thing moved with startling speed.

Through September and October the system was adopted by one metal-working factory after another, until by the end of October nearly all the main factories in Turin had workshop commissars and an internal commission elected by them.

On 31 October the first general assembly of Turin's workshop commissars met and adopted the Programme of the Workshop Commissars — the document Module 9 quoted, which put the election of commissars on universal suffrage of the workers at their workplace, by workshops. Note the sequence: the Programme codified a practice the plants had been improvising since August; it did not institute it. In roughly ten weeks the movement had gone from the first plant elections to a city-wide institution with a written programme — and these commissar-and-commission structures are what Turin, and this course from here on, calls the factory councils: Module 9's revolutionized internal commissions, in the flesh.

Hold that speed in mind, and hold it precisely. What spread was the commissar structure, through the metal-working factories — not necessarily the wider franchise, and not the whole industry of the city. That distinction matters, because Fiat-Centro had the structure too and produced the most conservative commission in Turin. The course does not know which franchise most of these plants used, and neither do you.

Even so, something real is being measured. A structure adopted plant by plant across the metal industry of Turin in ten weeks, ending in a city-wide assembly with a written programme, is not nothing. And it explains why the collision came when it did: a body covering the metal factories of Turin and claiming to speak for everyone employed in them is competing for exactly the ground FIOM held — and behind FIOM, the confederation that under the Pact of Alliance owned economic action.

Module 10 ended with a prediction: an institution claiming to speak for everyone in the metal plants of Turin could not be ignored by the unions whose ground it stood on. In the first two weeks of November 1919 the unions answered — locally, then nationally — and both sides' arguments survive in their own words. This module is the fight itself, and the union side deserves to be heard at full strength, because this course has so far mostly given you the council side's documents.

By late October the movement had real weight to throw: on 26 October an assembly of commissars at the casa del popolo of Turin represented 32 plants and 50,000 workers.

That was a preliminary gathering; the first general assembly, which adopted the Programme, followed on the 31st — Module 10's closing scene. The union's turn to answer came the very next day.

At the annual assembly of FIOM's Turin branch on 1 November, the leadership's spokesman was Uberti, and his argument was not a bureaucrat's reflex. It was a theory of what a union is, and Clark preserves it.

For Uberti the very existence of the union was at stake: extending the vote to non-union men was a negation of the trade union's raison d'être as a specialised organ of the class struggle and expression of the wishes of class-conscious workers.

Take the argument seriously, because Module 9 gave you the materials to. A union is a voluntary association: its members chose it, pay for it, and can be held to its discipline because they joined. That is precisely what makes it, in Uberti's terms, an organ of the class-conscious — the people who have already decided which side they are on. The council franchise hands a say over that instrument to men who have declined to join it: the indifferent, the Catholic, the syndicalist. On this view, "the electorate is the class" is not democracy but dilution — the conscious minority dissolved into a mass that has not chosen anything.

His practical proposal followed: recognize the commissars, but as consultative bodies only, under a general council of some 100 to 150 members elected by the union membership. The commissars could speak; the organized would decide.

Three motions stood at that assembly, and they map onto the positions this course has built. Uberti's: commissars as consultative organs of the union — bodies whose advice the union may take or ignore. A middle motion, based on a programme drawn up by the internal commission at Fiat-Centro — the conservative body Module 10 described — which would give the commissars a deliberative voice, a vote that counts, but exclude non-union men from electing them; it regarded the internal commission as nothing more than the representative of the trade union in the workshop. And the full council scheme, moved by Boero and Maurizio Garino — libertarian being the Italian movement's word for anarchist — carrying the Programme the general assembly of commissars had approved the previous day, on 31 October.

The full council motion was carried by a large majority, and a provisional executive committee of eleven was elected.

When the results were announced, the FIOM spokesman Castagno shouted ironically: “Viva l'U.S.I.!”

Decode the taunt with Modules 2 and 3 in hand — the doctrine from one, the federation from the other — because it is a precise accusation, not a noise. The USI was the revolutionary syndicalist federation: the current that held the class war should be fought on class terrain, inside industry, past the party and the established unions. Castagno was telling the winners what he thought they had just voted for: not a renewal of FIOM but the syndicalist heresy under a new name — the dissolution of the disciplined union into the undifferentiated mass. And an anarchist had just co-moved the winning motion, which gave the taunt something true to stand on.

Now notice the structural fact about that night, because it complicates both sides' stories. The assembly of 1 November was a union branch assembly, and a branch assembly's voters are its members — the class-conscious organized, in Uberti's own vocabulary.

That cuts two ways. Against Uberti: the dilution he feared was chosen, by a large majority, by exactly the electorate he said should decide — the organized voted to widen their own franchise. Against the councils: their first institutional victory was won inside and through the union machinery they claimed to supersede. The shape differs from Brevetti — there the union kept candidature under a universal electorate; here the electorate itself was the membership — but the pattern is the same in what matters: every recorded council victory so far ran through union-bounded channels, while the movement claimed a mandate from the class. The one election with a genuinely universal electorate, Brevetti's, is the one whose result was never recorded.

Togliatti, for the ordinovisti, addressed the adjacent confusion that same fortnight: Clark records him insisting that creating the councils and democratising the unions were two problems that should be kept separate. A council, in Module 9's terms, is a different kind of institution — not a more democratic FIOM. Whether the workers the commissars claimed to represent grasped that distinction is a question the sources do not answer.

A week later, on 9 and 10 November, the national FIOM leadership met in Florence, with four of the leading Turin council supporters — Garino, Boero, Sanmartino, Cerri — attending. The official view was put in a motion, and it passed unanimously among those voting, the four Turin representatives abstaining. It is worth reading clause by clause.

The motion began: the Union organisation must have entire responsibility for the movement and activity of the class, both within and outside the factory.

It warned of the dangers of new institutions superimposed on the union, which would thus come to be placed under the predominant influence of the non-Union masses.

And then it turned: the meeting decided to allow the experiment of Factory Councils — but only under conditions.

The conditions are the strategy, so take them one at a time. The councils' functions must be regarded as the continuation of the work of the internal commissions — which defines away Module 9's change of purpose. They must be co-ordinated with the union and based on its principles — which subordinates them to the body they claimed to outgrow. They must be restricted to centres containing the best union elements — and read that clause carefully, because it is shrewder than a ban: it concedes Turin, where the thing already existed among the strongest union branches in the country, and fences off everywhere else. A quarantine drawn around the outbreak. And their tasks must be defined so as to avoid the growth of facile illusions among the masses.

That last phrase should ring. It is the same phrase — at least as this course's translators render both documents — that the party used on 5 July 1919, warning the food insurrections of Module 6 against "facile illusions". Two different bodies, four months apart, reaching for the same words for enthusiasm they did not control. Not a ban, in either case. A permission so hedged that the permitted thing, done under the conditions, would no longer be itself.

The Florence motion codified a resistance that was already operating — the dates prove it, because the first rebuff outside Turin came a fortnight before Florence voted.

At the first congress of chemical workers at Milan, on 26-28 October, the Turin commissar scheme was discussed and rejected — the first impact of the Turin movement outside the city, and it was a rebuff.

In the same weeks — Williams's chronology places it in early November, without a precise date — Serrati, the effective leader of the party from Module 5, attacked the council movement in Battaglie Sindacali, the CGL journal Module 3 named. Be precise about what that alignment consists of: the metalworkers' federation resolving nationally against, and the party's effective leader attacking in the confederation's own paper. That is not yet the CGL as a body acting — but it is the machines of Part I visibly closing ranks, the reformist union leadership and the maximalist party leadership finding, for once, the same enemy.

So Part II's scoreboard at mid-November 1919 reads: the commissar structure through Turin's metal plants, on franchises the course cannot see; the full council scheme carried in Turin's FIOM branch; the national federation against; the party's leader against; the first export attempt rejected before Florence even met. Everything now turns on whether the councils can take the city's labour movement as a whole — the camera del lavoro, with every trade in it. That vote comes in December, and it is Module 12.

Module 11 left the councils holding the metal plants and one union branch, with the national federation, the party leader and the first export attempt all against them. December 1919 is when they take the city — the socialist section, then the camera del lavoro with every trade in it.

It is the high point of the councils as an institutional movement — the moment they hold the machinery of a city. (September 1920, when 400,000 to 500,000 workers occupied the plants, is a larger event by every other measure; whether the councils were what ran it is a question for Part III.) This module has an awkward shape because the sources do: the same weeks that produce the triumph produce the first clear evidence that it could not spread. Read the victory and the ceiling together, because they arrive together.

On 11 December the Turin section of the PSI voted overwhelmingly for the council system. Williams calls it easily won, and the shape of the win matters: the section passed under the leadership of an alliance between the ordinovisti and the abstentionists — the faction around Bordiga you meet properly in Module 13 — and sponsored a new Study Committee under Togliatti, with Boero, Tasca and Montagnana on it. It issued a proclamation to the Turin working class hailing its creative power.

On 11 December the Turin socialist section voted heavily to support the council movement, and an alliance of abstentionist and council communists took control of it.

Note the alliance, because it will not hold. The council communists had just won their political base by combining with a faction whose objection to the councils, when it comes, will be more damaging than anything the union leadership managed.

One piece of background belongs here, before the camera meets, because it arms the sharper of the two attacks the councils were about to face. The PSI had held its national congress at Bologna on 5-8 October 1919 — the congress where Serrati and the maximalists swept the board, and where the abstentionist fraction was formally organized.

At its Bologna congress of October 1919 the PSI committed itself to the creation of soviets in Italy — the workers' and soldiers' councils that had taken power in Russia in 1917.

That commitment mattered because it made the word contested property. Through the winter a whole series of academic soviet schemes was mooted inside the party — Bombacci's won majority approval at the national council in January 1920 — so by December every faction had a soviet project of its own to protect.

The camera del lavoro was the real prize, and what follows — the Turin camera vote — is the summit of the whole council movement. It was also a different order of difficulty. Module 3 defined it as the hub of every workers' organization in a district — not metalworkers but every trade, the institution Williams called far more significant locally than any British equivalent. It was summoned into special session on 14-15 December.

And the attack had a second prong. Beyond union prerogative, Serrati and the maximalists now charged that the council movement had falsely equated factory councils with soviets. The accusation was that Turin was dressing an industrial grievance body in the borrowed clothes of the Russian revolution, and it was shrewd precisely because the party had somewhere else it wanted the word to point.

Gramsci answered that charge by denying it — the councils, he said, were of absolutely original character and essentially political nature, which is a claim to have invented something rather than imported it. But in the same breath he gave ground on a different question, the one that mattered more.

Gramsci re-affirmed the councils' absolutely original character and essentially political nature, but admitted that in present circumstances their action could represent only an extension of trade-union activities — which would nevertheless radically alter their significance.

Sit with that concession, because it is the most honest sentence in Part II and it comes from the movement's own theorist at the moment of his greatest success. Everything Module 9 built — the change of purpose, the council as model of the proletarian state, the producer against the wage-earner — is here scaled back to an extension of trade-union activities, with the revolutionary content carried by the phrase "would nevertheless radically alter their significance." Williams's explanation is blunt: Gramsci was hard pressed given the actual practice of the movement in working-class reality.

Now be careful with the inference, because there is a tempting version of it that does not hold. Module 11 read the Florence motion's clause — councils as the continuation of the work of the internal commissions — as containment, defining away the change of purpose. Gramsci's concession is not quite a vindication of that clause, because he conceded an extension of trade-union activity, and Modules 7 and 9 were at pains to keep those apart: the commission is a grievance body inside one plant, the union a voluntary association bargaining across many. Fiat-Centro's commission, in Module 10, defined its job as servicing the union's agreements — commission work and union work pulling in the same direction there, but they need not.

What the concession does establish is the weaker and still damaging thing: whatever the councils were doing in December 1919, it was not yet the second change. They were operating on the wage-earner's side of Module 9's line, not the producer's. And that bears directly on the objection Module 9 told you to carry — that nobody had shown by what mechanism a body acquires a remit its contract denies it. Here is the movement's own theorist, at the summit of his success, testifying that it had not yet acquired one.

Having conceded the practice, the council communists won the vote anyway, and won large.

The council-communist argument, ably supported by the anarchist Garino, carried the day and won over even a majority of the delegates of Fiat-Centro.

That clause is the sweetest detail in Part II. Fiat-Centro is Module 10's plant: the one that kept the union-members franchise, returned no anti-union majority, and produced the most conservative commission in Turin. In December its own delegates crossed over.

A telegram from Serrati, urging the camera to leave the problem of votes for the un-unionized to the leading organizations of the PSI and the CGL, was overridden — and an ordinovista motion won by 38,000 votes to 26,000.

Two cautions on that margin. Sixty-four thousand votes were not sixty-four thousand people in a hall: as with Module 1's 591,245, these are delegate votes weighted by the membership each delegate spoke for — and the course cannot tell you which trades held which blocks, which is the fact you would most want. And the telegram and the motion are two events; Williams reports the telegram overridden and the motion carried, without saying the margin was a vote on Serrati personally.

Still, the shape of it is striking. The effective leader of the Italian Socialist Party intervened in a local labour assembly on the specific question of the franchise, and was ignored. A camera del lavoro is a union body, and under the Pact of Alliance the party leader had no standing there by treaty — so being overridden is the Pact working, not the party failing. The sharper evidence for functional centrism is three days earlier, on the party's own ground: the Turin section of the PSI had voted overwhelmingly against its own leader's line, and there he had every right to be obeyed and no means of securing it.

The camera's decisions were in theory binding on all working-class organizations, and it had called for the formation of councils throughout Italy.

So on paper this is total victory in the city: the party section, the metal federation's branch, and now the body whose word bound every workers' organization in Turin, calling for councils nationally. Williams describes what followed in a phrase worth keeping: by-passing the unions and their agreements with employers, the councils colonized the Turin complex.

Set that against the concession two cards back and you have a genuine contradiction — not the course's, but Williams's. A body whose action is only an extension of trade-union activities cannot simultaneously be by-passing the unions and their agreements with employers; the second is precisely the remit Module 9 said had never been shown to transfer. Williams asserts both within a page, and supplies no instance behind the second — no decision taken, no agreement broken, no case named. Prefer the concession, which is dated, attributed and made under hostile questioning; treat "colonized" as the historian's summary of a mood.

By the new year, as the first re-elections of workshop commissars went through, it was estimated that 150,000 workers had been organized in the council system.

Togliatti put the doctrine at its hardest in the journal's 3 January number, given over entirely to workers' control: “Between a bourgeois organization and an organization of workers there can be no compromise. There is no power to be shared between them. There is a power to be conquered.”

Hold that sentence next to the three preceding modules. No compromise — from a movement whose first success at Brevetti conceded a union monopoly of candidature, whose branch victory ran through union machinery, and whose theorist had conceded three weeks earlier that its action was an extension of trade-union activity. The doctrine is at maximum volume exactly where the practice is most compromised. That pattern has now recurred often enough in Part II to be a finding rather than a coincidence.

The group also built a school. On 30 November L'Ordine Nuovo opened a School of Propaganda with lecture and discussion courses: Gramsci on trade unionism and on soviets and the dictatorship of the proletariat, Terracini on factory councils, Togliatti on economic councils, Tasca on councils of soldiers and peasants. Zino Zini gave the inaugural lecture, and its title welds together the two transformations Module 9 kept separate — citizen into comrade, wage-earner into producer: From Citizen to Producer.

Now the turn, and Williams places it precisely at the top: it was at that moment of triumph that the journal began to reflect an awareness that its position was precarious. Three things had become visible.

First, the movement would not travel. An ordinovista mission to Milan in January 1920 achieved little; the articles on the city are unrelieved pessimism — a scattered working class, strong reformism, a loud and unorganized maximalism, and the syndicalists capturing the militants. The Turin council movement failed to extend itself outside Turin and its hinterland in Piedmont. Williams attributes this to the CGL, backed by the PSI, holding the line — and note that this is a stronger claim than Module 11 could make, where the evidence was one article in the confederation's paper rather than the confederation acting. His sharpest measure of the failure needs no such attribution: even within FIOM, Turin was isolated.

Second, the resistance was not only official. This is the finding that complicates Modules 10 and 11, and you should let it.

The ordinovisti's main preoccupation was the widespread popular revulsion against giving the vote to non-union men, which proved the major obstacle to council growth at the rank-and-file militant level — union leaderships proving more flexible, provided their ultimate control was ensured.

Read that against Module 11. The course has been presenting the franchise quarrel as leadership defending its prerogative against the shop floor. Williams says the opposite was the harder problem: the militants hated giving the vote to non-members, while the officials could be dealt with so long as control was preserved. The organized minority did not want to share, and it was not being ordered not to.

And now join that to the sentence in the paragraph above, because together they resolve something Module 11 left standing. On 1 November the organized militants of Turin's FIOM branch voted for the wider franchise by a large majority — which looked then like the shop floor overruling its officials. Williams's national finding says that is not what militants generally did. Put the two together and the reading changes: even within FIOM, Turin was isolated is not only a fact about geography, it is a fact about sociology. Turin's militants were the exception, not the rule, and Module 11's double edge should be read as local rather than general.

That costs Module 11's framing more than it first appears. The premise a council supporter had to reject — Uberti's equation of union membership with class consciousness — turns out to have been held by the organized workers themselves, not merely by officials defending a machine. Rejecting it did not put the ordinovisti on the side of the shop floor against the bureaucracy. It put them against both, with a constituency of the unorganized who have nowhere been shown to have wanted the vote.

Third, the other side had started organizing. Already in November 1919 — before the summit, which is worth noticing, as the Milan failure came after it — the Turin industrialists had begun to concert action against the councils, and the threat of an employer offensive grew.

One piece of bookkeeping before Part II's critics. Module 7 asked which way the commission would face when the interests of production and of the workers came apart. Four modules on, that question is still unanswered, and you should notice that this module has not answered it either. Everything here is votes, memberships, motions and estimates; not one instance of a council's actual conduct — no decision taken, no dispute handled, no case where output and the workforce pulled against each other. Gramsci's concession tells you why: in December the councils were still working the wage-earner's side of the line, where that divergence does not yet arise. The employers' offensive coming in Module 15 poses a different question again — whether the councils could survive an attack, which is not the same as whose side they take. Keep the two apart; the course will need both.

Module 12 ended with the councils holding Turin — and holding it in alliance with the abstentionists — Bordiga's abstentionist faction, to give it the name Clark uses — whose objection, that module warned, would be more damaging than anything the union leadership managed. This module is that objection.

Be clear first about how it differs from Module 11's. Those critics — FIOM nationally, Serrati, the Florence motion — were not merely defending turf: Module 11 was insistent that the union's case was a theory of what a union is, and that giving the vote to non-union men negated its reason for existing. But notice what even that principled objection grants. It argues over which body should hold which function in the workplace. It leaves the workplace standing as the unit on which the class is organized, and disputes the tenancy.

Amadeo Bordiga refused the unit. His case was that organizing the class by factory at all — whoever votes in it, whatever the body is for — builds the wrong thing, because the workplace is the wrong basis on which to gather a class. That is why Module 12 warned this objection would be the more damaging: the unions and the councils were arguing inside a premise they shared, and Bordiga denied the premise.

One instruction before you start, because this module will seem to contradict itself otherwise. Keep Bordiga's polemic and his structural objection in separate hands. The polemic is what he actually published against the men of L'Ordine Nuovo, and it misfired badly — a later card shows it landing on the wrong man. The structural objection is the argument about the workplace as a unit, and it is untouched by where the polemic landed. Judging the second by the failure of the first is the mistake this module exists to prevent.

Bordiga is the figure Module 12 promised, and he had been in this argument longer than anyone in it. A young engineering student in Naples, charismatic among socialist youth, he had been developing what Williams calls quasi-Leninist positions through the war years — that the proletariat must act politically and on its own account rather than in support of anyone else's cause, that the reformists had to be expelled, and by 1917 that the answer was armed insurrection — well before Comintern asked anyone in Italy for such things.

One detail dates the quarrel exactly. At the socialist youth congress at Bologna in September 1912, Angelo Tasca — the Tasca of Module 8, founder of L'Ordine Nuovo — proposed a radical change in the youth paper and was defeated by Bordiga. The two men had been fighting for seven years before there was a factory council to fight about, and you will want that fact at the card asking whose position he was actually hitting.

The base was small and it never stopped being small. Bordiga's influence ran through scattered groups of dedicated revolutionaries rather than any mass following, and Clark's verdict on the faction is blunt.

At the PSI congress at Bologna, 5-8 October 1919, Bordiga's abstentionist motion secured only 3,413 votes, against 48,411 for the maximalists and 14,880 for the centrist motion presented by Lazzari.

Hold that number against everything that follows: every argument in this module was made by a faction commanding about a twentieth of the congress. And read it the way Module 12 taught you to read the camera's 38,000 to 26,000 — these are weighted votes cast by delegates for the memberships behind them, not a count of heads in a room.

Bordiga's instrument was a journal, and its founding tells you how early he was.

Il Soviet came out in Naples in December 1918, calling for the expulsion of reformists from the decision-making centres of the party — long before any such call came from Comintern.

From January 1920 the journal ran three themes: a split in the party, abstentionism, and a critique of the factory councils.

Do not treat those as three separate campaigns that happened to share a masthead. Williams's point is that they "formed an integral unity, a total and totalizing perspective", and the unifying need was to define communism — to cut it absolutely free of the bourgeois world. Each theme is the same operation applied to a different surface: purge the party of those compromised by it, refuse the institutions that do the compromising, and reject the organizational form that reproduces bourgeois society's divisions inside the working class.

The tone is worth hearing, because it explains why the faction was easier to admire than to work with. Il Soviet, 1 February 1920, on the prospect of splitting the party: "Nothing's better than a good split... Above all, everyone in his right place... We'd know just who is a communist and who isn't; no more muddle... A good split lets the light in. Communists here, all the shower of opportunists over there."

Abstentionism needs defining precisely, because the obvious reading is wrong. It is not indifference to politics and not a judgement that elections are a waste of effort. It is the doctrine that the working class should refuse to contest elections to bourgeois parliaments at all — because participation is itself a source of the enemy's hold, teaching the class to seek its future through institutions built to contain it.

Note how neatly it locks to the split. If you renounce parliament and local government, you no longer need the men who are indispensable for that work — and those men, in the PSI, were precisely the reformists. Clark puts the logic plainly: expulsion of the right wing was a logical corollary of abstentionism. The tactic and the purge are the same move.

Bordiga's abstentionism began as an instrument — to break with parliament in order to clean reformism out of the party — but during 1920 it hardened into something approximating a fundamental principle.

That hardening — from tactic into principle — has consequences the course will need. By the fraction's national conference in May 1920 it was demanding an anti-electoral fraction inside the International itself. Williams judges that abstentionism proved a tough obstacle to any fusion of Bordiga's communists with the council communists — "less perhaps in itself than as a symptom of qualitatively different marxist sensibilities". The alliance that took Turin in December was built on a question neither side had settled.

Now the objection itself, which Bordiga had been making since September 1919. Its first move is a flat inversion of Module 9's claim.

Bordiga's position on the councils was that the party and political action were central and that council action was a diversion.

The reasoning underneath it: Module 9 built the council on the workplace — you are represented as a worker in a shop, by a commissar from your own reparto, because that is where production actually happens. Bordiga treated that same foundation as the flaw. The basis in craft and workplace, he argued, "set particular interests within the class against the interests of the class itself" — your shop, your trade, your plant, each with something to defend against the others. Interests of the class as a whole had to find expression in a form that transcended the particular, and that form was territorial — organized by the place people live, the district and the town, rather than by the plant they happen to work in. A ward contains the whole class: the engineer and the labourer, the employed and the sacked, the man at Fiat and the man at the tram depot. A workshop contains one slice of it, with its own interests to press.

For Bordiga soviets were territorial units, not units based on the productive processes of capitalism — and only soviets could be political.

Hold that against the definition Module 12 gave you, because they do not sit flush. There a soviet was the workers' and soldiers' council that took power in Russia in 1917 — bodies that plainly did grow out of workplaces and regiments. Bordiga is not reporting what the Russian soviets were; he is stipulating what a soviet ought to be, and using the Russian name for it. That is worth noticing, because the authority of the precedent is doing work in his argument that the precedent itself does not straightforwardly supply.

Read that against Module 12's victory and you can see how much it takes back. The Turin camera had just been captured for a scheme built on the productive process. Bordiga's answer is that the scheme was misconceived where it was strongest. He would allow the factory councils some technical role in the eventual socialization of production, and little more.

Bordiga put the underlying principle at its most general in an article of early 1920, and it is worth reading carefully, because it is easy to misapply. The revolutionary movement, he wrote, finds its true instrument in a body in which every individual enters as a member of the class, committed to a radical transformation of social relations, "not as a component element of a professional category, a factory or any local group whatever". And he names the body: as long as political power remains with the capitalist class, the general revolutionary interest of the proletariat can be represented "only on the political terrain, in a class party".

While political power remained with the capitalist class, Bordiga held that the general revolutionary interest of the proletariat could be represented only on the political terrain, in a class party — entered as an individual member of the class, not as a component element of a professional category, a factory or any local group whatever.

So hold Bordiga's two levels apart, because they answer different questions and you have now been given both. The party is how the class is represented now, while the capitalists still hold power, and you enter it as an individual who has transcended craft, trade and locality alike — which is what that clause is excluding. The soviet is the organ of power itself, and its basis is territorial rather than productive. The two do not collide: "any local group whatever" is a rule about what you may enter the party as, not a claim that a ward is the wrong shape for a soviet.

What both levels refuse is the same thing. A council organizes you as a member of a plant. Bordiga wanted the class organized as a class — in the party now, in territorial soviets afterwards — because he denied that a workplace is where a class acquires a general interest at all.

The second half of the critique is about timing, and it is the sharper half. On 21 September 1919 Bordiga stressed the absolute priority of the party and made the charge that gives this card its name.

To create organs of workers' power in factory councils while the bourgeoisie still held power was, for Bordiga, to make a formal imitation of a future institution — one lacking its fundamentally revolutionary character.

The argument is not that the councils were useless. It is that a form borrowed from the society you intend to build, and installed inside the society you intend to destroy, is a costume rather than an organ. What can genuinely represent tomorrow's proletariat today, he argued, is not a structure that mimics tomorrow but the workers who are conscious of the historical perspective — "workers inscribed in the communist party". The class struggling against bourgeois power is represented by its class party, "even if this is only an audacious minority", and — the conclusion that disposes of the whole Turin project — "the soviets of tomorrow must have their genesis in the local sections of the communist party".

That means that the thing Turin had spent a year building was growing from the wrong root. And it disposes of Module 10's franchise argument by refusing its terms: where Turin asked whether the vote should go to union members or to all workers, Bordiga was drawing his line somewhere else entirely, citing the Russian constitution to exclude anyone who lived on the labour of others — including, he noted, union men who also drew something from a scrap of land or capital, cases "not infrequent among us".

And by January and February 1920, with the split now his object, his general attitude hardened further: under bourgeois society the councils "could only serve reformist or syndicalist ends". Keep that sentence. It is the one the course has to test.

Here the module has to complicate the story it has just told, because Williams shows the critique missing its stated target. When Bordiga concentrated the fire of Il Soviet on "the Ordine Nuovo group", his real target was Tasca — the man he had beaten at Bologna in 1912. It was Tasca who wrote the major reply, Gradualism and Revolutionism in the Factory Councils, on 17 January 1920.

The tell is that Bordiga's own journal conceded the point. After he visited Turin, Il Soviet stated in February 1920 that Gramsci's ideas on the councils were quite different from Tasca's — an odd thing to announce about a group you have been attacking as a group.

Williams's judgement is that setting Bordiga's criticisms against articles specifically written by Gramsci is to take a trip through the looking-glass: they simply do not connect.

Why not? Because on the party, Williams finds, the two men's language is frequently identical — the need for a disciplined, centralized nucleus of militants in full awareness is Gramsci's position too. His summary: the "criticisms" of Il Soviet are, in the majority of cases, the "principles" of Gramsci.

Now apply the instruction from card 1, because everything above is about the polemic. What Williams has shown is that Bordiga's published attack on the Ordine Nuovo group failed to describe Gramsci: on the party, on discipline, on the centralized nucleus, the two were saying the same thing, so the attack hit Tasca and went past its stated target.

None of that touches the structural objection. "Organize the class by territory, not by workplace" is emphatically not among Gramsci's principles — he built the entire council theory on the workplace, on the reparto and the commissar and the productive process, and Module 9's whole case for why a council is not a union rests on it. That objection survives every word of this card, and Gramsci never answered it. So Module 12's promise stands: the damaging thing about Bordiga was never the polemic, which missed, but the angle it was fired from.

The critique found purchase for a second reason too, and Williams names it. The concentration through 1919 on building councils in Turin did produce a real obsession with the enterprise, and a genuine blur over what a soviet was — which is exactly the charge Module 12 recorded Serrati making.

And the charge had a formidable backer, which is worth knowing before Part III. The Comintern representative in Italy was Nicolai Ljubarsky, who wrote under the name Carlo Niccolini; with the weight of Moscow behind him he dismissed the factory committees, pointed to the undoubted fact that this form of workers' control had been short-lived in Russia itself, and hammered home the primacy of the unions. Williams's verdict is that his support was decisive in throwing the maximalists behind the CGL's resistance. So when Gramsci exploded in April — "No! No! comrade Niccolini has no right..." — he was not swatting a critic. He was losing an argument in Moscow.

The verdict on Bordiga's critique is where the mastery-level disagreement in this module sits, so take the two readings separately.

Clark measures influence and finds little. The abstentionists were disciplined and adequately organised but "too few in numbers to represent a serious threat to the leadership of the P.S.I." — 3,413 votes, scattered groups, no mass base. But he adds the qualification that makes them matter anyway.

Clark judged the abstentionists influential in Turin itself, and their uneasy relationship with the Ordine Nuovo group to have determined much of the future history of the Italian labour movement.

So on Clark's reading Bordiga is not a force in 1919. He is the reason the eventual communist party took the shape it did.

Williams measures the argument instead, and rejects the framing the course has inherited. Bordiga and Gramsci, he writes, were not "the Gog and Magog of Italian communism which 'history' has made them" — they were "twin if tautly dialectical polarities" of the same marxism. The real divergence is over theory and practice: Bordiga treats marxism as a science, with the risk that a science becomes a once-and-for-all possession and hardens into sectarian dogma; Gramsci treats it as history, with the risk of a historicism that dissolves communism into the reality it was supposed to master. Symptomatically, Bordiga saw Lenin as the restorer of marxism; Gramsci saw him as a creative innovator.

Williams does not stop at describing the two marxisms; he charges each man with the failure his own strength invited. Bordiga refused to recognize the Turin movement as one of his own historical situations. Gramsci failed to see the limitations of that movement as structural rather than contingent — as built into the design rather than the product of bad luck and union hostility.

Be careful with the word structural there, because it is not Bordiga's sense of it. The limitations Williams has in mind are the ones Module 12 set out — the movement would not travel beyond Piedmont, the rank and file balked at the wider franchise, the employers were already concerting — and his charge is that Gramsci read those as bad luck when they were built in. That is a claim about the movement's circumstances being permanent, not about the workplace being the wrong unit. It is not a vindication of Bordiga in different words.

What Williams does establish is narrower: each man was blind where his own strength lay, which is why his conclusion refuses to award the argument.

Williams concludes that it is difficult to see how an effective communist movement could be built on one to the exclusion of the other.

One last join, and it is the reason this module sits here rather than later — but get the order right, because the obvious version of it is wrong. Bordiga had been arguing since September 1919 that under bourgeois society a factory body could be no more than a technical or trade organ. In December 1919 Gramsci conceded that in present circumstances the councils' action could represent only an extension of trade-union activities. Only then, through January and February 1920, did Bordiga sharpen the charge to its hardest form: under bourgeois society the councils could only serve reformist or syndicalist ends.

So this is not a prediction meeting its confirmation. It is a running argument in which the other side's own theorist, at the moment of his greatest victory, said something that could be read as conceding the point — and the critic then hardened his claim. Whether it should be read that way is exactly what Module 12 told you to be careful about: Gramsci conceded an extension of trade-union activity, which is not the same as the internal commission work the Florence motion had tried to confine the councils to, and he attached to it the clause that the extension "would nevertheless radically alter their significance." Bordiga's "reformist or syndicalist ends" also reaches further than anything Gramsci said. The two sentences rhyme; they are not the same sentence, and the course has not earned the right to treat them as one. Part III inherits the question.

Module 11 gave you a shout. When the full council motion carried at the Turin FIOM branch assembly on 31 October, the union spokesman Castagno answered the result by shouting ironically "Viva l'USI!" — an accusation, as that module said, and a precise one: it told the winners that the leadership now thought it was fighting syndicalism.

This module asks whether the accusation was true, and the answer turns out to be more interesting than yes or no. The syndicalists and anarchists were not the councils' enemies. They were among their few friends — and possibly, though this is contested, among their inventors.

The guide here is Carl Levy, whose study of Turinese anarchism joins Spriano, Williams and Clark among this course's main sources — and is the first of them to put the libertarians rather than the socialists at the centre of the story. His organising claim is that Gramsci's council communism did not spring only from the war and its aftermath: the ideas of self-management and an autonomous working-class culture were present in pre-war Italy already, and when council communism emerged it bore the marks of that inheritance — while Gramsci remained critical of the same tradition's instability and demagoguery.

Levy's name for the anarchists is Gramsci's other fratello-nemico — brother-enemy — the first being Bordiga of Module 13.

The phrase is Levy's, not the course's, and both halves of it earn their place. Brother, because — as this module will argue — the councils grew in ground the libertarians had prepared. Enemy, because Gramsci spent years polemicising against them, and because their objection to organizing production by technical function bites on precisely the body Module 9 described. Set that beside Module 13's brother-enemy: Bordiga denied that the workplace could be the unit on which a class is organized at all, where the anarchists disputed how a workplace body should be run and who should have to join a union to matter in it.

Module 2 defined revolutionary syndicalism and Module 3 gave you the USI that carried it. Neither defined anarchism, and the course has been using the two words as near-neighbours ever since. They are not the same, and the difference is about what a union is for after the revolution rather than before it.

For the anarchists, trade unions served only to ameliorate the conditions of the working class and, in a crisis, to mobilize workers for revolutionary action.

That is a limited job, and deliberately so. Syndicalism, as Module 2 had it, wants the union to become the structure of the new society — the class war fought on class terrain, by the organization of the trade. Anarchism refuses to give any single institution that role.

The anarchists envisaged a post-revolutionary society built on a plurality of organizations based on free accord, the unions being only one such institution.

And there is a second split, which explains a good deal of the friction inside the USI itself. Syndicalism inherited anarchism's hostility to the state, but it kept insisting on the need for experts and functional hierarchies to run production — and that insistence, Levy writes, disturbed many anarchists. Hold that beside Module 9's council, which allotted representation by craft in proportion to each craft's contribution to the product. A body organized by technical function is a syndicalist sort of body, and some anarchists were uneasy with it on principle. Keep that unease: it is the whole of the enemy in brother-enemy, and this module will come back to it. The anarchists' quarrel with the movement was not about whether the councils should exist — it was about how a workplace body should be constituted, and about who had to hold a union card to count.

Turin anarchism was small and mattered anyway, and the clearest evidence for this module's claim is not in Levy at all. It is in what you were already told. In Module 11 the full council scheme at the FIOM branch was moved by Boero and Maurizio Garino — and the course noted then that Garino was an anarchist. In Module 12 the council-communist argument at the camera del lavoro was ably supported by the anarchist Garino, and carried.

Put that plainly, but do not overstate it, because the temptation here is real. Module 12 counted three institutional victories in Turin — the party section, the FIOM branch, and the camera del lavoro — and an anarchist is on the winning side of two of them. The third, the section on 11 December, fell to an alliance of ordinovisti and Bordiga's abstentionists with no anarchist in it at all. Two of three is still a striking figure for a movement Levy calls decidedly minority; it is not the same as saying the anarchists built the councils, and this module will not say that.

What it does establish is narrower and enough: when the council system won its institutions, libertarians were among the people carrying the motions. Whatever Castagno meant by his shout, he was not describing a movement the syndicalists and anarchists were fighting.

Levy names Maurizio Garino and Pietro Ferrero as the two outstanding Turin anarchists — skilled industrial workers and astute labour organizers who earned the respect of radical and reformist socialists alike, and of the rural arrivals in Turin's expanding factories.

Levy's judgement is that if anarchism became a viable, if decidedly minority, movement in Turin, it was largely because of the prestige those two earned. Note what kind of prestige it was: not a newspaper's and not a theorist's, but the standing of skilled men on the shop floor — the same ground the workshop commissars of Module 9 were elected from.

Now the correction this module exists to make. The anarchists' quarrel in these years was not with the factory councils. It was with FIOM, and the issue was one Module 7 has already given you a name for.

Garino held that FIOM's most controversial decision was its closed-shop policy — which offended the non-unionized and, many radical FIOM members thought, needlessly alienated sympathetic migrant workers.

The principle underneath the complaint is exactly anarchist. For Garino and his comrades, Levy writes, joining a trade union was meant to be a free decision — otherwise the ideals of the labour movement would be compromised by a corporate mentality. Compulsion may deliver members; it cannot deliver the thing membership was supposed to signify.

Read that against Module 10 and the alignment becomes obvious. The whole Turin argument about the franchise — union members only at Fiat-Centro, the whole workforce voting at Brevetti — was an argument about whether the union card should be the gate. The anarchists had been against that gate for years, on grounds that had nothing to do with Gramsci. On the question Module 10 said the councils turned on, they were already on the councils' side.

The plain fact about the libertarian left and the council movement is the opposite of what Castagno's shout implies.

Williams records that every variety of libertarian and syndicalist welcomed the councils, and that the Programme of the Workshop Commissars was openly friendly to the syndicalist federation USI.

Anarchist militants were active inside the movement — at Savona in Liguria they had real influence among anarchists and socialists both — and the USI's standing grievance in these months was against the metalworkers' national agreement, whose factory regulations the councils were widely said to have by-passed — and Module 12 told you to hold that claim loosely, since Williams asserts it without naming a single agreement broken. What matters here is not whether it was true but that the USI believed it and was pleased: the syndicalists were cheering the councils for the very thing the union federation accused them of. That is the shape of it: the syndicalists' fight was with the union, alongside the councils.

Which leaves Castagno's accusation looking odd, and worth restating precisely. When the union leadership called the councils syndicalism, it was not describing an alliance that existed on the ground so much as naming the heresy it feared. The charge tells you what the accuser thought, and Module 11 already said that was a fact worth having. It does not tell you what the councils were.

Here the module reaches a genuine historiographical question, and it will not be settled — but you should be able to state both sides, because what turns on it is whether council communism was Gramsci's invention or the movement's inheritance.

It has been suggested that the anarchists themselves advanced the proposal to transform the internal commissions into factory councils as early as October 1918.

Against that stands a piece of evidence which cuts against the interest of the side giving it. In December 1919 the anarchist De Bartolomeis wrote that the idea of the factory council was culled from British and Russian examples, as filtered through L'Ordine Nuovo — an anarchist, that is, crediting Gramsci's journal rather than his own comrades.

A third witness muddies it further. A month earlier the local reformist Mario Guarnieri had claimed that in the debates before the previous autumn's Rome congress, socialist intellectuals — Gramsci and his friends — helped the opposition to the reformists inside the Turin branch of FIOM, on a platform of greater organizational democracy, opposition to collaboration with the state, and proletarian unity. The historian Emilio Soave replies that Guarnieri was misrepresenting the facts in order to serve his own 1919 quarrel with the ordinovisti — which is a charge about motive, and does not by itself dispose of the claim.

What tips Levy's judgement is not the testimony but the timing, and one organization.

The Fascio Sindacale d'Azione Rivoluzionaria was, in Levy's assessment, the most serious attempt at a united front in 1919.

Levy notes that L'Ordine Nuovo was launched just as individual maximalist socialists in Turin were seeking a united front with the libertarians, and that the leading radical pressure group of 1919 — the Study Committee for Proletarian Unity, which is not the Study Committee Module 12 put under Togliatti in December, a different body sharing two words — resembles the later ordinovista groupings in how it was put together. On balance he judges the connection closer than is usually believed.

Which is the answer to this module's title question, so state it carefully. Not that the anarchists invented the factory council; the evidence does not carry that, and one of their own denied it. What the evidence does carry is that the council was assembled in a city where the libertarian tradition had spent a generation arguing for self-management, free association and shop-floor autonomy, by people who were talking to that tradition while they worked. Brother, then, as well as enemy.

One structural point remains, and it is the one that costs the movement most.

The USI stood outside the Pact of Alliance, which bound only the PSI and the CGL — so the federation most certain to vote for insurrection was excluded from the September 1920 vote by the machinery's own design.

Module 3 built the workers' side out of two machines bound by the Pact of Alliance — a treaty between the PSI and the CGL, dividing political action from economic. A treaty has signatories, and the USI is not one of them. Nothing in that constitution has a place for a federation claiming 800,000 that belongs to neither party nor confederation, which is why Module 1 could tell you the USI was excluded from the September 1920 vote — the wing of the movement most certain to vote for insurrection was, by the machinery's own design, not in the room.

Keep that for Part III, where it becomes expensive: in September the USI will press for immediate occupation and be overruled by bodies it holds no vote in. And keep Garino's name for the module after next, because when the Turin metalworkers elect a Committee of Agitation — a strike leadership elected by the commissars themselves rather than by the union — in the crisis of March 1920, he is on it.

Two modules of critics have argued about what the councils were. Bordiga said they were built on the wrong unit; the anarchists said they were welcome but the union card should not be the gate. Both arguments are internal to the movement, and both assume the councils will still be there next month.

This module is the other side of the table deciding they should not be. It also answers the question Module 12 left open — not whose side the councils take when production and the workforce pull apart, but whether they can survive being attacked, which is a different question and turns out to be the urgent one.

One definition first, because the whole module turns on it. A lock-out is the employers' converse of a strike: the owners shut the plant and keep the workforce out, paying nobody, until terms are accepted. Where a strike withdraws labour from the employer, a lock-out withdraws the workplace from the workers.

Now see why that is the precise instrument against the institution Part II has been building. A trade union survives a lock-out: its members are enrolled by trade, its funds sit in a bank, its officials keep their desks, and it can bargain from outside the gates. A factory council cannot. Module 9 built it on the workplace — commissars elected by reparto, representation apportioned by each craft's share in the product — so a council whose factory is shut is cut off from the thing that constitutes it. It cannot meet where it meets, it governs nothing, and its members are outside the gate as individuals.

A union survives a lock-out because it is enrolled by trade and can bargain from outside the gates; a factory council is weaker against one, because everyone who elects it and everything it governs are inside the plant.

Do not overstate it, though — Module 16 will lock these councils out and find them still deliberating three weeks later. A lock-out does not abolish a council. It suspends the ground it stands on, which is enough to make it the employers' natural weapon and not enough to finish the argument.

That is worth holding beside Module 13. Bordiga argued the workplace was the wrong basis on which to gather a class, and meant it as a claim about representation: a body built on a plant speaks for a plant. The lock-out is a different objection with the same target — that a body built on a plant can be separated from it. The two are not the same argument, but the employers were about to give the workplace basis a test Bordiga never had to make.

The employers' national body has been in this course since Module 1, where its spokesman told the workers it was their turn to drop their trousers, and Module 3, which gave it a founding. In the spring of 1920 it becomes the decisive actor.

Williams's glossary has Confindustria revived in March 1920 to embrace regional and sectional consortia such as AMMA, the Association of Metallurgical, Mechanical and Affiliated Industrialists, based in Turin — and calls it a very powerful and aggressive organization.

Here the course owes you a disagreement rather than a fact, and Module 5 gave you the method for handling one: say whose claim each is. Williams dates the body's first formation to 1910 and calls March 1920 a revival. Clark says the General Confederation of Industry was founded in March 1919, and calls Gino Olivetti its Secretary-General where Williams calls him director. Nothing in this course turns on which is right; what matters is the thing both agree on, which is that by March 1920 the employers had a national organization capable of coordinating a response and the movement did not.

Set that against Module 3's inventory. The workers' side entered 1920 as a party and a confederation bound by a treaty, with a syndicalist federation outside it and a Catholic one beside it. The employers entered 1920 with one body, one secretary, and a membership that could be instructed.

Before the man, the map, because this module will name four employers' bodies and they are easily confused. Clark lays them out as a nesting of three levels. The Turin Industrial League is the local one — by 1906 a more overtly political organisation than its predecessors, taking in 75 employers across industries and bargaining collectively. AMMA is the regional layer, the metal and engineering employers of the province. Confindustria is the national body: the war's growth and concentration of industry, Clark writes, made employers realise the need for a national organisation.

Now the man, because he is all three at once. Gino Olivetti — a brilliant lawyer, in Clark's description, who gradually extended the League's scope — was secretary of the Turin Industrial League, ran the regional employers' associations in engineering and vehicle-making, and became Secretary-General of the national confederation, keeping his Turin post until mid-1920. In November 1919 he also became a deputy. He worked closely with Giovanni Agnelli, the founder of Fiat.

That is the asymmetry to hold, and state it carefully. It is not that the workers' side had no continuous personnel — D'Aragona and Buozzi have been in this course since Module 3. It is that on the employers' side one man held the local, the regional and the national office simultaneously, and sat in parliament. Module 3's workers' side put its jurisdictions in different hands and wrote a treaty to divide them.

What he was answering matters as much as what he said. Gramsci had just published an essay closing on a programmatic demand — the next step after the councils was to be a first national congress of factory councils, to which every factory in Italy would be summoned, "represented by delegates elected directly and not by trade-union officials". That is the council claim at its widest: a national body of the whole class, assembled around the union rather than through it.

The day after Gramsci's article appeared, Olivetti declared war on dual power in the factory — and Williams adds that the climax of the council movement was also its crisis.

Dual power names a situation in which two authorities claim the right to decide the same thing in the same place and neither can abolish the other — the classic case being a government and the revolutionary councils sitting beside it.

Be careful about whose word it is, though, because it changes what the card can claim. Dual power is a revolutionary term of art: Spriano attributes it to Lenin, and it belongs to the vocabulary of the people Olivetti was fighting. Williams reports the declaration in those words; whether Olivetti chose them or Williams supplied them, the sources here do not settle.

What survives either way is the shape of the charge. Olivetti's complaint is not indiscipline or breach of agreement — that is the language he uses to the prefect, four cards from now, and it is a different accusation. Here what is being declared war on is an authority in the workplace, which is what Module 9 said a council was. Whether or not the phrase is his, the employers were treating the council's own claim about itself as true, and acting on it, before the movement had settled whether to believe it.

The next three cards happen in a room, and you need to know whose room. A prefect is the central government's own officer in a province — appointed from Rome, not elected locally, responsible for public order and answerable to the Ministry of the Interior. He commands the police, can ban meetings, and can call on troops.

Module 1 has already shown you what that office can do, though it did not name it then: when the factories were occupied in September 1920 and no policeman was arguing, the police were absent by order, because prefects and chief constables were following precise instructions. The prefect is where a decision taken in Rome becomes a policeman standing still.

A prefect is the central government's appointed officer in a province, responsible for public order and answerable to the Ministry of the Interior; in March 1920 the prefect of Turin was Taddei.

What follows is known because Taddei reported it to the Ministry of the Interior the same day, and Clark found the despatch in the state archives. It is the clearest view this course gets of the employers and the state in a room together.

On 20 March 1920 Olivetti, De Benedetti of the local Industrial League and Agnelli of Fiat told the prefect of Turin that the industrialists had decided to have recourse, within a very brief period, to a general lock-out.

Their stated grievance, in Taddei's summary, was that indiscipline and the workers' continual exorbitant demands in the factories had reached such a point that a general lock-out was now necessary. This was not an ultimatum to the unions and not a public announcement. It was private notice to the state, in advance, of an act that would put tens of thousands of men into the street.

Taddei's answer is the hinge of the module, and it is not the one you would expect from either the employers' account or the movement's.

He protested. A general lock-out, he said, was a danger to public order. But in the same breath he pledged the support of the public authorities to the industrialists if they chose instead to apply the existing factory regulations rigorously, and declared his readiness "on all occasions to intervene in order to repress any violence that might occur, or any criminal attacks on their rights."

The industrialists therefore agreed to give up their plans for a lock-out — and warned that they would apply the proper disciplinary sanctions in every case where the regulations or agreements were not observed. Read that as an exchange, because that is what it was: they surrendered the blunt instrument and were handed a licence to use a fine one, with the state standing behind it.

Because no existing agreement mentioned factory councils, the industrialists had in effect secured State backing for their aims, knowing the government would intervene to prevent any occupation of their factories.

That clause about the agreements is the sharpest thing in the module, so take it slowly. Module 6 taught you the February 1919 contract — the eight-hour day, the Commission of Six, the fineable unofficial strike. Module 9 taught you that the council was a body the contract does not name. Both facts were, until now, arguments about legitimacy. In the prefect's office they become an operational advantage: rigorous enforcement of the agreements is a policy that cannot protect an institution the agreements do not mention. The councils were about to be attacked through a rulebook that had no word for them.

One further document shows the employers acting as a disciplined body rather than a collection of firms, and it is worth having because the movement never managed the equivalent.

Olivetti's confidential circular of April 1920 instructed member organizations that if local political authorities pressed for concessions on the councils, it was their duty to refuse co-operation and warn the Confederation at once.

The instruction anticipates that local political authorities — mayors, councils, the socialist communes of Module 6 — might press employers to concede on the councils, and forbids members to deal with them independently. The employers had understood something the movement had not: that a concession granted in one city sets a rate for every other.

Module 1 said that telling an absent state apart from a collapsed one is the hinge of this whole course, and offered September 1920 as the case. Here is an earlier and smaller case, and it does not fit either description.

The state in Taddei's office is not absent: it is present, attentive, and makes promises. It has not collapsed: it is functioning exactly as designed, receiving information and issuing assurances. But neither is it neutral between the parties. It has undertaken, privately and in advance, to put its police behind one side's reading of a contract and against any occupation by the other.

Clark judges that neither the prefect nor the central authorities had any awareness of the nature of the factory council movement: they regarded the unrest at Turin as purely a matter of public order, which it was their duty to preserve.

Resist the conclusion that this makes the state the employers' instrument, then, because that judgement of Clark's blocks it. A government that does not know what a factory council is cannot be conspiring against one. It is something more useful to the employers than an alliance and harder for the movement to answer: a state that will reliably do what public order requires, in a situation where one side has arranged for public order to require exactly what it wants.

Module 20 will put the same state in front of a far larger occupation and ask why it did nothing, and the answer you accept there should be consistent with what it did here.

Module 15 ended with a licence: the employers gave up their general lock-out and took, in exchange, the state's promise to back a rigorous enforcement of the existing agreements. This module is that licence being spent, starting within the week.

It is also the end of Part II, and the module the whole part has been pointing at. By 24 April the council system still exists on paper and has been destroyed in substance, and the manner of its destruction determines what happens in September.

The trigger was, in Clark's words, a banal issue: the introduction of summer time on 21 March 1920. Note that the two sides were moving independently and almost simultaneously — the employers were in the prefect's office on 20 March, and the Chamber of Labour issued its manifesto against the coming clock change the same day. Neither was responding to the other. What the employers had bought was a way of answering whatever came next, and this is what came next.

You need one piece of context to see why a clock mattered. Legal time — putting the country's clocks forward by decree — had been a characteristic feature of the wartime Industrial Mobilisation of Module 4. Reimposing it in 1920 was read as the government's wish to keep wartime industrial conditions going.

The Turin Chamber of Labour called summer time intolerable to proletarian sentiments, mindful of the dictatorial decrees of the war period, and on 20 March issued a manifesto telling workers to ignore it.

So for a few days Turin ran on two times at once: industry on solar time, the schools and every state institution on legal time. The dispute is not really about an hour. It is about whether the war is over.

On 22 March the internal commission at Fiat's Industrie Metallurgiche works asked the management to put back the hands of the factory clock, which had been altered even though inside the plant work was proceeding on solar time. The director refused.

When the director refused to put the factory clock back, the internal commission asked a clock-maker to do it — and the following day its members were dismissed for insubordination.

Read the dismissal against Module 15. This is the enforcement policy the employers bought on 20 March, applied three days later: not a lock-out, not a confrontation over pay, but the rigorous application of factory discipline to a body whose standing the agreements never guaranteed.

The workers answered in a way that decides how you read September.

The workers of Industrie Metallurgiche answered the dismissals with an internal strike — they stopped work, but remained inside the factory.

That is the occupation of the factories in miniature, five months early and in one plant. Remember it when Part III opens: the tactic that astonished Italy in September 1920 was already being used in March, over a clock. (Whether Turin invented it is another matter — see the card on the escalation.)

And be precise about which body acted, because it is not the one you might expect. The dismissed body was the internal commission — but Module 10 established that by the end of October 1919 nearly all the main Turin factories had workshop commissars and an internal commission elected by them. So this is not the commission acting instead of the council. It is the council system acting through the organ Module 7 introduced.

A second conflict opened the same week, and it answers a question Module 3 has kept open since the Pact of Alliance.

At the Acciaierie Fiat works the men struck on 23 March when their demand that internal commission members be paid for the time spent on their duties was refused. Clark notes what made it more than a pay dispute: the workers failed to use the stipulated bargaining machinery whereby the local branch of FIOM negotiated on their behalf, and in that sense their action was a clear breach of the national agreement.

At Acciaierie Fiat the men struck on 23 March without using the bargaining machinery through which FIOM's local branch negotiated for them — a clear breach of the national agreement.

So the shop floor is now going around its own union. Module 3 gave you the jurisdictional question as a quarrel between two machines at the top of the movement. Here it arrives from below, and the body being bypassed is FIOM.

Clark's summary of both disputes: they were regarded by workers and employers alike as matters of principle, in both of them the employers' authority within the workshop was the issue at stake, and both demonstrated the industrialists' new-found determination to insist on the rigorous application of national agreements — in accordance with the understanding reached between them and the prefect of Turin.

The bargain of Module 15 is now redeemed, and the paperwork survives.

On 24 March the prefect promised the industrialists he would clear the workers out of the Industrie Metallurgiche works, and was backed the same day by the Minister of the Interior.

The Minister's instructions were not narrow. He urged the prefect to forbid public meetings, expel all suspicious foreigners, arrest known trouble-makers, and generally "inspire the conviction that the Government intends to guarantee order and calm, and take bold measures". Clark's judgement is that the prefect's support for the employers was, in fact, a crucial factor in the situation.

The workers were cleared from the plant by police and troops on the evening of 24 March, at Agnelli's request. Taddei's own dispatch to Rome records the request and his answer: Agnelli asked him to clear the factory, "which had been illegally occupied by the workers, who were holding meetings there but not working"; he gave the order, and added — "I renew my urgent request for ten officers trained in street fighting methods, because it is probable that tomorrow work will be abandoned in all factories and there will be a general strike."

Module 15 read the state from a private conversation in which nothing had yet happened: a prefect who protested a lock-out and then promised to back rigorous enforcement. Here is the same reading tested against troops in a factory, and it holds — with one thing added that the conversation could not show.

What was added is the scale of what the state was willing to do while believing it was doing something small. Clark's judgement was that the authorities had no awareness of the nature of the factory council movement and treated Turin as a question of public order. On 24 March that produced a plant cleared by police and troops, a ministerial instruction to ban meetings and arrest trouble-makers, and a prefect requisitioning street-fighting officers — all of it, in the state's own mind, routine order-keeping. A government does not need to understand an institution to destroy it, and this one did not.

Negotiations over the dismissals failed on a point of principle rather than money. Management would rescind the dismissals — but insisted that the members of the former commission should not hold office again for a year.

The commissars read that as an attack on the principle that "the Internal Commission emanates from the workers alone, without interference from the industrialists". Gramsci wrote that the proposal would deprive the commission's members of their proletarian civil rights. The employers also demanded that in future the commission's functions be restricted to the limited ones laid down in the national agreement.

On 26 March FIOM's Turin branch convened the workshop commissars of every metal plant, who elected a Committee of Agitation — the anarchist Garino among its members — and called an internal strike that began on 27 March in 44 of the city's engineering factories.

Note who is leading: not union representatives but a committee elected by the commissars themselves. And note that the tactic may not have been Turin's invention: Clark suggests it was borrowed from a Ligurian example, where syndicalist-led plant occupations had run since February.

On 29–30 March the employers answered with a lock-out, enforced by troops. At the Diatto Frejus works two rows of barbed wire were strung to keep the workers out, and Avanti! wrote that there were enough troops in Turin to "quell an entire region in revolt".

Stop on that, because it settles what Module 15's bargain was worth. The industrialists had given up a general lock-out on 20 March in exchange for state backing. Nine days later they had the enforcement and the lock-out, with troops the same state supplied. They surrendered nothing. What the trade bought them was a first move that put the state on their side of it — and the state, having promised to keep order, could hardly refuse the troops once disorder followed.

Here is the deepest reason the councils could not be defended, and it is not about force at all.

Negotiations were formally always about the powers of the internal commissions, the only bodies recognised in factory regulations; factory councils and workshop commissars were never mentioned, which made a negotiated settlement on that issue impossible.

The Industrial League made it a constant theme of its propaganda that the negotiations were not about factory councils. It could say so honestly. No agreement mentioned them; Module 9 built an institution the contract had no word for, and Module 12 recorded that nobody had ever shown by what mechanism a body acquires a remit its contract denies it.

Clark adds the detail that makes the position stranger still: the workers' main aim was to defend an unofficial institution that had been bitterly attacked by the unions. They were striking, against employers and state, in defence of a body their own federation had spent Module 11 trying to contain.

The dispute nearly ended without a general strike, which is worth knowing before you read the strike as inevitable.

On 2 April the workers' representatives conceded almost every specific: that the commission had no business altering the clock, that its members should resign, that the clock should show legal time, that the workers should not be paid for the hours of stoppage, and that they should have consulted FIOM before striking. They held only the principle.

Then, with the whole thing apparently settled, the employers introduced new proposals at the last minute — norms to regulate and discipline the internal commissions, which the commissars read as withdrawing the recognition the councils had been tolerated under. The prefect wired Rome on 11 April that "some difficulties have arisen at the last minute".

The workshop commissars rejected the employers' final proposals despite Buozzi's pleas that they were a big advance on the existing official regulations.

Bruno Buozzi is Module 3's able FIOM leader, and he was probably right on the paperwork: the terms did improve on the official regulations. He was arguing about a document. The commissars were defending an institution the document did not recognise, which is why an improvement in the document could not settle it.

The escalation to a general strike came through the institution Module 12 spent its whole length capturing.

On 13 April 1920 the Turin Chamber of Labour proclaimed a general strike to defend the factory councils, and it spread across Piedmont: Williams puts half a million workers out, in defence of what he calls the first affirmation of the communist revolution in Italy.

Answer the jurisdictional question Module 3 trains you to ask. Not the CGL, which owned economic strikes by treaty. Not the party, which owned political ones. The camera del lavoro — the city body the councils had taken in December, now spending the victory of Module 12 on the crisis of Module 16. The single most important thing the council movement ever won is being converted, four months later, into its defence.

Solidarity decided the outcome, and the story is about the rest of the movement rather than about Turin.

The PSI and CGL pledged to call a general strike if the government used force against the Turin workers. Beyond that, Williams writes, the socialist movement as a whole "turned a face of brass". The Milan edition of Avanti! would not even print the Turin strike bulletin. The national council meeting was moved to Milan — a city in the grip of a general strike, Gramsci commented acidly, not being a suitable theatre for socialist debate.

Williams records that during the April strike the syndicalists were the only ones to move: railwaymen at Pisa and Florence refused to carry troops to Turin, and dockers at Genoa and Livorno struck in the same cause.

Hold that against Module 14. The federation outside the Pact of Alliance, the one the movement's constitution had no room for and the September vote would exclude, is the one that stopped trains and docks for the councils. Liguria was convulsed; the Genoa syndicalists called for a general strike throughout Italy. A delegation from Genoa travelled to Milan to press for it and was told the strike was being satisfactorily settled.

The decisive meeting was the PSI national council at Milan, and it is the fullest test yet of Module 2's charge.

Turin sent Tasca and Terracini. Tasca argued that in the face of an alliance of capitalists and the state the CGL and the party must move, and called for a general strike. Terracini went further: a general strike was no longer enough, and "the party must annul the Pact of Alliance and take over the struggle". In three months, he said, all Italy would be in Turin's situation — "We say, prepare for insurrection!"

After the Turin delegates called for a general strike and for insurrection at the PSI national council of 19-21 April 1920, not a single voice was raised in support.

What the council did instead is the detail to remember. Its principal agenda item was Bombacci's scheme for soviets, and it spent interminable hours on the feasibility of that. They went on chattering about soviets and councils, Gramsci snarled, while in Piedmont and Turin half a million workers starved to defend the councils which already existed.

Two more things happened at that council, and each is worth carrying forward.

Togliatti brought to Milan a document you will meet properly in Part III — For a Renewal of the Socialist Party, the ordinovisti's programme for remaking the PSI, which Moscow would praise later that year — and presented it. It was ignored. And Bordiga — the critic of Module 13, whose objection was that the councils were built on the wrong unit — attacked the leadership for its irresolute behaviour while offering Turin, in Williams's words, nothing but principled doctrinal criticism. Williams calls this the moment Bordiga demonstrated the limits of his doctrine and his imagination.

The Turin men, in despair, withdrew their own motion. That left two. The leadership proposed one which — rich in maximalist rhetoric, in Williams's phrase — called for more time to instil communist principle and to prepare a proletarian armed force. Against it stood a motion from Misiano, which Williams calls equally empty but rather more urgent in tone.

The Directorate's motion carried by 61,000 votes to Misiano's 26,000, and Bordiga's group — some 7,500 votes — abstained. It was a disavowal of Turin.

Set the result beside Module 2. The charge was that a leadership of this kind, facing a real chance of action, will re-affirm the maximum programme and decline to act on it. Here it is at full size: asked to move while half a million workers were already out, the party voted for more time to instil communist principle and for an armed force to be prepared at some later date. Note that the minority was not Turin's either — 26,000 went to a motion Williams calls equally empty. Almost nobody in the room was voting for the thing that was actually happening.

The end came through the machinery, not the troops.

The Turin Study Committee issued a last call for an Italian general strike and the strike leaders approached Errico Malatesta, the most celebrated anarchist in Italy and no part of the socialist machinery at all — a measure of how far outside their own institutions the Turin leaders had been driven. But D'Aragona of the CGL, the man Module 1 showed moving the winning motion in September, moved remorselessly in to settle it. By 24 April it was over.

The councils survived April 1920 but were emasculated: in particular they were not to operate during working hours, and remained as shadows, their power merely potential.

Look hard at that clause, because you have met it before. Module 6 taught you the February 1919 agreement, which recognised the internal commissions and then added that they were to fulfil their tasks only outside working hours. Fourteen months and one general strike later, that is exactly where the councils end up. The April settlement does not invent a restriction; it restores the one the movement had already signed.

The most consequential damage was not to the councils but between the parts of the movement, and it is the reason this module sits at the end of Part II.

Employer authority was re-established and union influence restored; Williams judges that the Gramscian-communist perspective was eliminated and the council movement itself to a degree discredited. A terrible gulf of hatred opened between the Turin left, the ordinovisti in particular, and the CGL leadership with which the party Directorate was associated.

And one fact arrives here that this course has not shown you being made, which is itself worth noticing. Williams records that the ordinovisti had built an organization in the rural areas — the countryside beyond Turin, where Module 3's land-workers were. Nothing in Part II described them doing it; the story you have been told is a story about metal plants in one city, and Module 13 even recorded Williams faulting the movement for not travelling beyond Piedmont. Whatever the rural network was, the course meets it at the moment of its destruction.

After April the union men uprooted the organization the ordinovisti had built in the rural areas so effectively that during the September struggle nothing the Turin men said was believed there.

That is the sentence Part III has to be read against. When September comes and Turin cannot carry the country, the reason is not only Module 12's geography. Some part of the answer is that a rural network had been destroyed in the spring by the movement's own confederation, and the Turin men's credibility with it.

Take stock before Part III, because the occupation is about to be fought by a movement in the condition this module has just described.

Module 7 asked which way the internal commission would face when the interests of production and of the workers came apart. April answers it: dismissed en bloc for moving a clock, defended by an internal strike, and fought over until the plant was cleared by troops. Whatever the commission had been in 1906, by 1920 its constituents treated it as theirs and the employers treated it as a rival authority.

The April general strike ran from 13 to 24 April 1920 and ended in defeat: Williams records the Turin communists broken by a lock-out supported by 50,000 troops.

But the council system that Modules 8 to 12 built is, by 24 April, a shadow with potential power, confined to non-working hours, disavowed by its party, contained by its union, and cut off from the countryside. Everything Part III describes — the wage claim, the occupation, the vote in Milan — happens to a movement that has already lost this argument. That is what Module 21 will have to explain.