Leon Trotsky

The Chinese Revolution

The Permanent Revolution
& the Canton Insurrection


Source: The Militant, Vol. IV No. 33 (Whole No. 92), 28 November 1931, pp. 4 & 2.
Transcription/HTML Mark-up: Einde O’Callaghan for the Trotsky Internet Archive.
Copyleft: Leon Trotsky Internet Archive (www.marxists.org) 2013. Permission is granted to copy and/or distribute this document under the terms of the Creative Commons Attribution-ShareAlike 2.0.

The following is a chapter, The Permanent Revolution and the Canton Insurrection, from the book on the Chinese Revolution to be issued in the near future by the Pioneer Publishers. We reprint this chapter now by L.D. Trotsky because of the general interest in the subject, and particularly because December marks the anniversary of the Canton Insurrection. This material has never before appeared in the English language. Further information on the comprehensive book on the Chinese Revolution will be forthcoming in the next few days.Ed.

In November 1927, the plenum of the Central Committee of the Chinese party decided that

“The objective circumstances existing at the present time in China are such that the duration of a directly revolutionary situation will be measured not by weeks .or by months, but by long years The Chinese revolution has a lasting character, but on the other hand, it has no stops. But its character, it constitutes what Marx called a permanent revolution”.

Is this right? Intelligently understood, it is right. But it must be understood according to Marx and not according to Lominadze. Bucharin, who showed up the latter precisely for having employed this formula, was no closer to Marx than the author of it. In capitalist society, every real revolution above all if it takes place in a large country, and more particularly now, in the imperialist epoch, tends to transform itself into a permanent revolution; in other words, not to come to a halt at any of the stages it reaches, not to confine itself to national bounds, but to extend and to deepen itself up to the complete transformation of society, up to the final abolition of class distinctions, consequently, up to the complete and final suppression of the very possibility of new revolutions. That is just what the Marxian conception of the proletarian revolution consists of, being distinguished by that from the bourgeois revolution, limited by its national scope as much as by its specific objectives.

The Chinese revolution contains within itself tendencies to become permanent in so far as it contains the possibility of the conquest of power by the proletariat. To speak of the permanent revolution without this and outside of it is like trying to fill the cask of the Danaides. Only the proletariat, after having seized the state power and having transformed it into an instrument of struggle against all the forms of oppression and exploitation, in the interior of the country as well as beyond its frontiers, gains therewith the possibility of assuring a continuous character to the revolution, in other words, of leading it to the construction of a complete socialist society. A necessary condition for this is to carry out consistently a policy which prepares the proletariat in good time for the conquest of power. Now, Lominadze has made of the possibility of a permanent development of the revolution (on the condition that the broadest policy be correct) a scholastic formula guaranteeing at one blow and for all time a revolutionary situation “for many years”.

The permanent character of the revolution thus becomes a law placing itself above history, independent of the policy of the leadership and of the material development of revolutionary events. As always in such cases, Lominadze and company resolved to announce their metaphysical formula relative to the permanent character only after the political leadership of Stalin, Bucharin, Tchen Du-Siu [1] and Tang Ping Shan had thoroughly sabotaged the revolutionary situation.

After having assured the continuity of the revolution for many years, the plenum of the Central Committee of the Chinese Communist Party, freed from any further doubts, deduces from this formula conditions favorable to the Insurrection.

“... Not only is the strength of the revolutionary movement of the toil ing masses of China not yet exhausted but it is precisely only now that it is beginning to manifest itself in a new advance of the revolutionary struggle. All this obliges the plenum of the Central Committee of the Chinese Communist Party to recognize a directly revolutionary situation exists today (November 1927) throughout China.”

The Objective Conditions on the Eve of the Insurrection

The Canton insurrection was deduced from a similar evaluation of the situation with a perfect inevitability. Had a revolutionary situation really existed, the mere fact of the defeat of Canton would only have been a special episode, and in any case, would not have transformed the uprising of this city into an adventure. Even in face of unfavorable conditions for the insurrection at Canton itself or its environs, the leadership had as its duty to all that was necessary to realize the revolt most rapidly in order thus to disperse and weaken the forces of the enemy and to facilitate the triumph of the uprising in the other parts of the country.

However, not after “many years” but after a few months, it had to be acknowledged that the political situation had declined abruptly, and that before the Canton insurrection. The campaigns of Ho Lun and Yeh-Tin were already developing in an atmosphere of revolutionary decline, the workers were separating themselves from the revolution, the centrifugal tendencies were gaining in strength. This is in no way contradictory to the existence of peasant movements in various provinces. That is how it always is.

Let the Chinese Communists ask themselves now: Would they have dared to decide upon fixing the Canton insurrection for December had they understood that for the given period the fundamental forces of the revolution were exhausted and that the great decline had commenced? It is clear that if they had understood in good time this radical break in the situation, they would in no case have put on the order of the day the appeal for the armed uprising in Canton. The only way of explaining the policy of the leadership in fixing and carrying out this revolt, is that it did not understand the meaning and the consequences of the defeats of Shanghai and Hupeh. There can be no other interpretation of it. But the lack of understanding can all the less excuse the leadership of the Communist International, since the Opposition had warned in good time against the new situation and the new dangers. It found itself accused for this by idiots and calumniators of having the spirit of liquidators. The resolution of the Sixth Congress confirms the fact that an inadequate resistance to “putschistic moods” produced the fruitless uprisings of Hunan, of Hupeh, etc. What is to be understood by “putschistic moods”? The Chinese Communists, in conformity with the directions of Stalin and Bucharin, judged that the situation in China was directly revolutionary and that the partial revolts had every chance of being extended successfully to the point of becoming a general insurrection. In this way, the launching of these surprise attacks resulted from an erroneous estimation of the circumstances in which China found itself towards the second half of 1927, as a result of the defeats suffered.

The Duplicity of the E.C.C.I.

In Moscow, they could prattle about the “directly revolutionary situation”, accuse the Oppositionists of being liquidators, while providing for themselves beforehand against the future (especially after Canton) by making reservations on the subject of “putschism”. But on the theater of events, in China itself every honest revolutionary was duty bound to do everything he could in his corner to hasten the uprising, since the Communist International had declared that the general situation was propitious for an insurrection on a national scale. It is in this question that the regime of duplicity divulges its deliberately criminal character.

At the same time the resolution of the Congress says:

“The Congress deems it entirely inexact to attempt to consider the Canton insurrection as a putsch. It was a heroic rearguard (?) battle of the Chinese proletariat, fought in the course of the period which has just passed in the Chinese revolution; in spite of the crude mistakes committed by the leadership, this uprising will remain the standard of the new Soviet phase of the revolution.”

Here confusion reaches its zenith. The heroism of the Cantonese proletariat is placed in evidence as a screen to cover up the faulty leadership not of Canton (which the resolution casts off completely) but of Moscow, which only yesterday spoke not of a “rearguard battle” but of the overthrow of the government of the Kuo Min Tang.

Why is the appeal to insurrection denounced as putschism after the experience of Canton? Because thanks to this experience, the inopportuneness of the uprising was confirmed. The leadership of the Communist International had need of a new lesson by example in order to discover what already appeared quite clear without it. But are not these supplementary lessons for behind-handed people, given in life, too costly to the proletariat?

Lominadze, one of the infant prodigies of revolutionary strategy, swore at the Fifteenth Congress of the Communist Party of the Soviet Union that the Canton insurrection was necessary, right and salutary, precisely because it inaugurated an era of the direct struggle of the workers and peasants for the conquest of power. He met with agreement. At the Sixth Congress, Lominadze recognized that the insurrection did not inaugurate an era of triumph but concluded one of defeat. Nevertheless, just as before, the uprising is considered necessary, right and salutary. Its name has simply been changed: from a clash between the vanguard of the forces at hand, they made a “rearguard battle”. Everything else remains as in the past. The attempt to escape the criticism of the Opposition by hiding behind the heroism of the Cantonese workers has as much weight as, let us say for example, the attempt of General Rennenkampf to take shelter behind the heroism of the Russian soldiers whom he drowned by his strategy in the Masurian swamps. The proletarians of Canton are guilty, without having committed mistakes, simply of an excess of confidence in their leadership. Their leadership was guilty of having had a blind confidence in the leadership of the Communist International which combined political blindness with the spirit of adventurism.

The False Comparison Between Moscow 1905 and Canton 1927

It is radically false to compare the Canton insurrection of 1927 with that of Moscow in 1905. During the whole of 1905, the Russian proletariat rose from one plane to the other, wresting concessions from the enemy, sowing disintegration in its ranks, concentrating around its vanguard ever greater popular masses. The October 1905 strike was an immense victory, having a world historical importance. The Russian proletariat had its own party, which was not subordinated to any bourgeois or petty bourgeois discipline. The self-esteem, the intransigence, the spirit of offensive of the party, rose from stage to stage. The Russian proletariat had created Soviets in dozens of cities, not on the eve of the revolt but during the process of a strike struggle of the masses. Through these Soviets, the party established contact with vast masses; it registered their revolutionary spirit; it mobilized them. The czarist government, seeing that each day brought a change in the relationship of forces favorable to the revolution, passed over to the counter-offensive and thus prevented the revolutionary leadership from being able to gain the time needed for continuing to mobilize its forces. Under these conditions, the leadership could and should have staked everything so as to be able to test by deeds the state of mind of the last decisive factor: the army. This was the meaning of the insurrection of December 1905.

In China, the events developed in a directly opposite way. The Stalinist policy of the Chinese Communist Party consisted of a series of capitulations before the bourgeoisie, accustoming the workers to support patiently the yoke of the Kuo Min Tang. In March 1926, the party capitulated before Chiang Kai-Shek ; it consolidated his position while weakening its own; it discredited the banner of Marxism; it converted itself into an auxiliary instrument of the bourgeois leadership. The party extinguished the agrarian movement and the workers’ strikes by putting into practise the directions of the Executive Committee of the Communist International on the bloc of the four classes. It renounced the organization of Soviets so as not to disturb the situation at the rear of the Chinese generals. It thus delivered to Chiang Kai-Shek the workers of Shanghai, bound hand and foot. After the crushing of Shanghai, the party, in conformity with the directions of the Executive Committee of the Communist International, placed all its hopes in the Left Kuo Min Tang, the so-called “center of the agrarian revolution”. The Communists entered the Wuhan government, which repressed the strike struggle and the peasants’ uprisings. They thus prepared a new and still crueller devastation of the revolutionary masses After all this, an instruction entirely penetrated with the spirit of adventurism, was issued, ordering an immediate orientation towards the insurrection. It is from this that was first born the adventure of Ho Lun and Yeh-Tin, and the even more painful one of the Canton coup d’état.

No, all this does not resemble the insurrection of December 1905 at all.

An Inopportune Insurrection

If an opportunist calls the events of Canton an adventure it is because it was an insurrection. If a Bolshevik employs the same designation for these facts, it is because it was an inopportune insurrection. It is not for nothing that a German proverb says that when two men say the same thing it does not mean the same thing. The officials à la Thälmann can continue, on the subject of the Chinese revolt, to recount to the German Communists the “apostacy” of the Opposition. We will know how to teach the German Communists to turn their backs on the Thälmanns. In actuality, the question of evaluating the Canton insurrection is the question of the teachings drawn from the Third Congress, in other words, of a lesson where the life of the German proletariat was at stake.

In March 1921, the Communist Party of Germany sought to engage in an insurrection by basing itself upon an active minority of the proletariat in the face of the passive spirit of the majority, which was tired, distrustful, expectative, as a result of all the preceding defeats. Those who directed this attempt at this time also sought to take shelter behind the heroism of which the workers gave proof in the March battles. However, the Third Congress did not congratulate them for this attempt when it condemned the spirit of adventurism of the leadership. What was our judgment in those days of the March events?

“Their essence,” we wrote, “is summed up in the fact that the young Communist party, alarmed by a manifest decline in the workers’ movement, made a desperate attempt to profit by the intervention of one of the most active detachments of the proletariat in order to ‘electrify’ the working class and, if possible, to bring matters to a decisive battle.” (L. Trotsky, Five Years of the Communist International, page 333)

Thälmann has not understood a thing of all this.

From July 1923 on, we demanded, to the great astonishment of Klara Zetkin, Varski and other old, very venerable but incorrigible social democrats, that the date of the insurrection in Germany be fixed. Then, at the beginning of 1924, when Zetkin declared that at that moment she envisaged the eventuality of an uprising with much “more optimism” than during the preceding year, we could only shrug our shoulders.

“An elementary truth of Marxism says that the tactics of the socialist proletariat cannot be the same in face of a revolutionary situation as when this situation does not exist.” (Lenin, Works, Vol. XV, page 409)

Today, everybody acknowledges this ABC verbally, but how far they still are from applying it in reality!

The Tasks of the Leadership

It is not a question of knowing what the Communists must do when the masses are rebelling of their own accord. That is a special question. When the masses arise, the Communists must be with them, organizing and instructing them. But the question is posed differently: What did the leadership do and what should it have done during the weeks and months that immediately preceded the Canton insurrection? The leadership was duty bound to explain to the revolutionary workers that as a consequence of defeats, due to an erroneous policy, the relationship of forces had veered entirely in favor of the bourgeoisie. The great masses of workers who had fought tremendous battles, dispersed by the encounters, abandoned the field of battle. It is absurd to believe that one can march towards a peasant insurrection when the proletarian masses are departing. They must be grouped together again, fight defensive battles, avoiding a general battle, which obviously does not hold out any hope. If in spite of such a work of clarification and education, contrary to it, the masses of Canton had rebelled (which is very unlikely) the Communists would have had to put themselves at their head. But it is just the reverse that happened. The uprising had been commanded in advance, deliberately and with premeditation, based upon a false appreciation of the whole atmosphere. One of the detachments of the proletariat was drawn into a struggle which obviously held out no hope, and made easier for the enemy the annihilation of the vanguard of the working class. Not to say this openly, is to deceive the Chinese workers and to prepare new defeats. The Sixth Congress did not say it.

Does all this signify that the Canton insurrection was only an adventure, allowing of but one conclusion, that is, that the leadership was entirely incompetent? No, that is not the sense of our criticism. The Canton insurrection showed that even after enormous defeats, with the manifest decline of the revolution, even in non-industrialized Canton, with its petty bourgeois traditions of Sun Yat Senism, the proletariat was able to rise in revolt, to fight valiantly and to conquer power. We have here a fact of enormous importance. It shows anew how considerable is the weight of the proletariat in its own right, how great is the political role which it can eventually play, even if the working class is relatively weak in numbers, in a historically backward country, where fhe majority of the population is composed of peasants and scattered petty bourgeois. This fact, once more after 1905 and 1917, completely demolishes the philistines à la Kuusinen, Martinov and consorts, who teach us that one cannot dream of speaking of the dictatorship of the proletariat in “agrarian” China. Yet the Martinovs and the Kuusinens are at the present time the daily inspirers of the Communist International.

The Canton insurrection showed at the same time that at the decisive moment, the proletariat was unable to find even in the petty bourgeois capital of Sun Yat Senism a single political ally having a distinct form, not even among the debris of the Kuo Min Tang, of the Left or the ultra-Left. This means that the vital task of establishing the alliance between the workers and the poor peasants in China devolves exclusively and directly upon the Communist party. The accomplishment of this task is one of the conditions for the triumph of the coming third Chinese revolution. And the victory of the latter will restore the power to the vanguard of the proletariat, supported by the union of the workers and the poor peasants.

Rebuild the Chinese Communist Party

If “apostacy” must be spoken of, the traitors to the heroes and the victims of the Canton insurrection are those who seek to rid themselves of the teachings of this uprising in order to conceal the crimes of the leadership. The lesson to draw is the following:

  1. The Canton insurrection showed that only the proletarian vanguard in China is capable of carrying out the uprising and of capturing power. The revolt showed, after the experience of collaboration between the Communist party and the Kuo Min Tang, the complete lack of vitality and the reactionary character of the slogan of the democratic dictatorship of the proletariat and the peasantry, opposed to the slogan of the dictatorship of the proletarian drawing the poor peasants behind it.
  2. The Canton insurrection, conceived and executed contrary to the course of development of the revolution, accelerates and deepens the decline of the latter, facilitating the annihilation of the proletarian forces by the bourgeois counter-revolution. This stamps the inter-revolutionary period with a painful, chronic and lasting character. The greatest problem now is the renascence of the Communist party as the organization of the vanguard of the proletariat.

These two conclusions are equally important. It is only by considering them simultaneously that the situation can be judged and the perspectives fixed. The Sixth Congress did neither the one nor the other. By taking as its point of departure the resolutions of the ninth Plenum of the Executive Committee of the Communist International (February 1928) which assured us that the Chinese revolution “is continuing”, the Congress slipped up in its flight to the point of declaring that this revolution has now entered into a preparatory phase. But this flight will not help anything. We must speak clearly and sincerely, recognize firmly, openly, brutally the breach that has taken place, adapt the tactics to it and at the same time follow a line of conduct which leads the vanguard of the proletariat through the insurrection to its preponderating role in the Soviet China of the future.


1. Tchen Du-Siu since then has acknowledged the incorrectness of his position, and accepted the viewpoint of the Left Opposition. His statement has been printed in the MilitantEd.

return return return return return

Last updated on: 11.2.2013