A persistent confusion holds that, for the Fourth International, any nationalisation of industry made a “workers’ state”. The confusion has disoriented generations of Trotskyists. It led them to defend regimes that had nothing to do with working-class struggle, and to lose the analytical precision that separated revolutionary Marxism from campist apologetics. Ted Grant’s theory of Proletarian Bonapartism is its best-known source. But it would be too easy, and wrong, to say that Grant invented it. The truth is more awkward and more useful.
The Fourth International’s own 1951 resolution on Eastern Europe defined the new states above all by their economic base, and said plainly that they had arisen without the revolutionary action of the masses. What it also did was attach conditions to that conclusion. Grant dropped them. So, as it happens, did some of his sharpest critics. Recovering those conditions is the way out of the confusion, and the subject of this editorial and the next.
What the Fourth International said in 1951
The Third World Congress met in Paris in August 1951. Its resolution on the class nature of the countries of the Soviet buffer zone was carried by 41 votes to 2. It said three things that matter here.
First, the buffer states had become deformed workers’ states “above all” because of their economic base: nationalised and planned property relations essentially like those of the USSR.
Second, those states had arisen “not from the revolutionary action of the masses but from the military-bureaucratic action of the Soviet bureaucracy”, in exceptional circumstances created by the war.
Third, and this is the sentence later quoted without its qualifiers, the resolution said that mass revolutionary action had turned out not to be an indispensable condition for the bureaucracy to destroy capitalism. It added that this held only in analogous exceptional conditions, and in an international climate like that of the Cold War.
The resolution added an explanation that is easy to miss. Even here, the bureaucracy did not do without the masses entirely. It mobilised them bureaucratically, in committees of various kinds that helped disarm and expropriate the bourgeoisie. And it had to, the resolution said, because the bureaucracy is not an autonomous social force, a class. It leans partly on the proletariat to fight the bourgeoisie, while locking the masses into its own police control.
China was not covered. The resolution lists China separately, next to the USSR and the buffer zone, and the Congress Manifesto of the same summer was still calling on Chinese workers to finish destroying bourgeois power. The Fourth International began calling China a workers’ state only in 1952–55, and its American co-thinkers in the SWP waited until 1955.
This was not the position Ernest Mandel had started from. In 1946 he had argued that the bureaucracy could not fully assimilate a territory without calling on the autonomous action of the masses, which it feared like the plague, so assimilation would stay limited. By January 1951, in his Ten Theses, he had conceded. His own account of his error was candid: what he had misjudged was not the bureaucracy’s wish to assimilate the buffer but its capacity to do so without the masses outflanking it. The workers stayed passive, and the Kremlin managed. He kept one reservation. Property relations, he wrote, “isolated from their historical origin” do not by themselves settle a state’s class character.
Nor did he drift from that position later. In his Critique de l’eurocommunisme (1978), Mandel set out his settled view of how the buffer states were made. A bureaucracy whose privileges rest on a mode of production born of the abolition of capitalism can, occasionally and within narrow limits, try to extend its power by extending the zone where that mode of production operates. It did so in 1939–40 in eastern Poland, the Baltic states and Bessarabia, and on a larger scale in 1947–49 in its military glacis in Eastern Europe. Each time it acted by military-bureaucratic means under its own strict control, led no genuine mass popular revolution, and had a prior agreement with imperialism on the division of the world into spheres of influence. He added that nothing of the kind should be expected again in Europe or Asia in the foreseeable future. That is a description of an exception, bounded by its conditions, not a general rule about nationalisation.
So the Fourth International’s 1951 position was not “mass mobilisation or nothing”. It was narrower and more defensible. Capitalism had been abolished from above, by a bureaucracy that was not a class, under conditions specific enough to be named.
What Grant did with it
Grant was a serious Marxist whose contributions deserve respect: his analysis of post-war capitalism, his understanding of reformism, the organisational tenacity that carried his tendency through decades on the margins. His theory of Proletarian Bonapartism began as a genuine contribution, first applied to China in 1949. In the colonial world the national bourgeoisie was often unable to complete the democratic revolution, and the proletariat was too small, or too badly led, to take power. Grant argued that this double incapacity gave a military-bureaucratic caste room to act as a third force, balancing between the classes, and that once the Soviet model existed, such a caste could expropriate capital and plan the economy.
The move that did the damage was to make property the whole test. A state’s class character, Grant held, is fixed by the form of property it defends: arguing against Tony Cliff in 1949, he explained the USSR’s class nature by the form of property alone, and in 1978 he put the rule in a sentence, that the economy of a class determines its class character. On that test, how a regime came to nationalise does not enter the question. That deletes exactly what the 1951 resolution had specified. Gone are the analogous conditions, the Cold War, a bureaucracy that was not a class and had to lean on the workers, and a ruling apparatus pushed into expropriation against its own programme. What was left was a formula: nationalisation plus planning equals a deformed workers’ state. Once the conditions were gone, nothing stopped the formula travelling far beyond the cases it had been built for.
The Burma problem
In March 1962 General Ne Win took power in a coup and nationalised large sectors of the Burmese economy under the banner of the “Burmese Way to Socialism“. There was no mass mobilisation, no destruction of the old state in revolutionary struggle, no working-class agency of any kind. The army took over the existing state and decreed state ownership.
Grant’s framework, applied consistently, made Burma a deformed workers’ state, and his tendency said so. In “The Colonial Revolution and the Deformed Workers’ States“ (Militant International Review no. 14, 1978), Grant tells the story approvingly. The army, tired of the landowners’ and capitalists’ failure to hold the country together, and with China next door as a model, mounted a coup, expropriated them and proclaimed a “Burmese Buddhist Socialist State”. He then puts Burma first in a list, with Syria, Angola, Mozambique and Ethiopia, of countries where, with Cuba and China as models, social relations had been transformed. A year later the same journal described the state factions in Cuba, Syria and Burma as having turned to abolishing private ownership, “leaning on the workers and peasants”, under “a dictatorial regime of proletarian Bonapartism”. In 1986 it named Burma, with Ethiopia, Syria, China and Vietnam, as a model on which new deformed workers’ states might be set up. By property relations alone, Burma looked like Cuba or China, and the tradition treated it accordingly.
This was absurd. Burma’s junta had no relationship to working-class struggle. It suppressed national minorities, jailed dissidents, and enriched the officer corps. Its nationalised economy was a means of military enrichment, not a form of transition towards socialism.
Restoring the conditions
Calling Burma a dictatorship is not enough. Eastern Europe was a dictatorship too. The question is what separates the buffer zone, Yugoslavia, China and Cuba from Burma and Ba’athist Syria, when every one of them nationalised under an unelected apparatus. The 1951 conditions, stated with the actors named, answer it in four steps.
1. A coalition meant to restabilise capitalism. In each real case the ruling party began by trying to put capitalism back on its feet, not to abolish it. That is how the people’s democracies of 1944–47 worked, with bourgeois ministers kept in office under Communist control of the police. So did Mao’s “New Democracy” and its “bloc of four classes”. So, in its way, did the July 26 Movement’s democratic programme. Moscow wanted capitalism in the buffer, so as to secure the USSR through peaceful coexistence.
2. A class pressure that made restabilisation impossible. That pressure could come from imperialism, from the masses, or from both. In Eastern Europe it was mainly imperialist: the Marshall Plan and the Cold War turned the surviving bourgeois parties into a bridgehead for rollback. Imperialism’s role was double, as Mandel noted in 1978. It had already conceded the region as a Soviet sphere of influence, which is why it did not intervene. Inside that sphere, its Cold War pressure made a capitalist buffer a threat rather than a safeguard. In China it was the American drive to the Yalu and the blockade. In Yugoslavia it was a partisan war, factory occupations, and the rapid collapse of the coalition. In Cuba it was Washington’s hostility and a mass movement together. A government of this kind, sitting on a bourgeois state, cannot stand still. It goes forward to expropriation or back to capitalism. Finland and Austria went back. So did Italy and France in May 1947, when the same Stalinist strategy ended with the Communists put out of government. The parties’ intentions were the same in East and West. The class relation was not.
3. A ruling apparatus that was not a class, forced against its own programme, and leaning on the workers. Expropriation was not what these leaderships set out to do. They were pushed into it. To carry it out they had to mobilise the working class, however bureaucratically: committees, militias, campaigns against the capitalists. That is the 1951 resolution’s point. A bureaucracy that is not a class has to borrow the social weight of one.
4. Expropriation of the capitalist class, on the model of the existing workers’ state. The bourgeoisie was destroyed as a class, not just stripped of titles. The economy was reorganised on the only non-capitalist model in existence, inside a bloc that depended on it. A peasantry, however mobilised, cannot supply a state form of its own. That was Trotsky’s argument, and Mandel restated it about China in 1952.
Burma fails at the first two steps, and Grant himself supplies the evidence.
No restabilisation was defeated. There was no coalition of the 1944–47 kind, set up to restore capitalism and then pushed off course by a class it could not contain.
No class pressure forced the turn. The pressure Grant describes is rebellion and the ruling class’s incapacity, the ordinary crisis of a weak post-colonial state.
It was the caste’s own choice. Grant explains why officers in such countries switch to “proletarian Bonapartism”: it enlarges their power, prestige, privileges and income. Nationalisation was not forced on the junta against its programme. It was the programme, adopted because it suited the officer corps, and Ne Win intended exactly what he did.
The third step needs more care than “leaning on the workers” suggests. The Militant tradition did say Burma’s regime leaned on the workers and peasants, and it is true that the junta built workers’ and peasants’ councils. But it built them from above, after the expropriations, as instruments of the new single party’s control. That is bureaucratic mobilisation as a means of rule. It is not the borrowing of a class’s weight to defeat a bourgeoisie the regime could not otherwise defeat, against a programme it had set out with. What distinguishes Eastern Europe, Yugoslavia and China is not that the masses were organised, since every one of these regimes organised them, but that the ruling party was pushed into expropriation against its own intentions.
Step 4 was only half met: state ownership without a non-capitalist bloc behind it, and without the bourgeoisie destroyed as a class, as distinct from the Indian and Chinese merchant layers the regime pushed out. Syria fails the same way. The nationalisations there were the Ba’ath’s own instrument of rule, never a retreat forced on it by a class it could not otherwise contain.
That also explains the different endings. A ghost ship can be turned round at will, because no class put it to sea. When Burma’s generals decided in 1988 that markets served them better, the “Burmese Way” was dropped by decree. No social force had made it and none could stop its reversal.
The Grant tradition’s reply
Defenders of Grant will say that this puts the subjective factor above objective social relations. They will argue that workers’ interest in defending a state depends on the property relations it upholds, not on the manner of its birth.
The four steps are not a subjective test. Every one names an objective class relation: the coalition’s programme, the pressure that defeated it, the social force the apparatus had to lean on, and what was done to the capitalist class. The objection also misses where the real difference lies. When a military caste nationalises industry while keeping wage labour for profit, joining the world market on subordinate terms and using the state to enrich itself, it has not abolished capitalism. It has concentrated capitalist relations in the hands of the state.
The consequences
Dropping the conditions has had lasting consequences.
It led to the “defence” of regimes with no relationship to working-class interests. Trotskyists who called for the unconditional defence of Burma or Syria as “deformed workers’ states” lined up with military dictatorships that suppressed workers, oppressed minorities, and enriched ruling cliques.
It confused what workers actually have an interest in defending. Workers defend the gains of a real overturn, the expropriation of the capitalists and the planned economy, even when a parasitic bureaucracy administers them. They have no interest in defending a junta’s state enterprises, which exist to exploit them.
And it helped prepare the ground for the campist drift. If the test for “defence” is nationalisation plus hostility to Washington, any regime that passes it can be defended. That is the logic behind much of the British left’s echoing of the Morning Star, and for groups like the CWI and RCI it has roots in Grant’s formula.
Restoration
Grant’s framework also struggled with what actually happened to the states it described. It expected political revolution or a capitalist restoration imposed from outside. It did not expect restoration by the bureaucracy itself. In 1989–91 the ruling stratum decided private property suited it better, and turned itself into a bourgeoisie while the workers watched.
The four steps do not make the real cases safe from that, and should not be read as if they did. What they explain is the difference in how the reversal came. Where an overturn was forced by class pressure and carried out by leaning on the workers, reversing it took a political and social crisis of the whole bloc. Where it was a junta’s decree, the junta’s next decree was enough.
Next
The question is not “has property been nationalised?” It is whether a class pressure defeated a project of capitalist restabilisation, and forced an apparatus that was not a class to expropriate the capitalists by leaning on the workers. The next editorial develops that test as the theory of Conditional Transitional Society. Conditional because the conditions were in the Fourth International’s own resolution all along. Grant’s error was to drop them, not to depart from the Fourth International.
Note: This post was updated in October 2026 to make better use of an archive we have built of Militant International Review. We now cite MIR 14, 17 and 32 to more clearly make Grant’s point that the caste acted on its own interest and “the workers were organised, but not borrowed.”