J.P. Cannon on leadership election

Richard Fidler rfidler at cyberus.ca
Wed Nov 20 14:42:18 MST 2002

The revolutionary regroupment process now under way in Australia and parts
of Europe, and possibly soon in other areas of the world (Latin America?),
poses afresh a number of organizational questions within the new emerging
formations. Among these are or will be the need to select leadership bodies
within these formations and their appropriate mode of election. This is an
important issue for any political formation that seeks to play a day-to-day
leadership role in the class struggle in all its forms.

The issue has been alluded to recently on this list by a number of comrades:
Bob Gould, Philip Ferguson, Louis Proyect among others. Lou in particular
proferred the view that seats on the U.S. SWP national committee, in his
day, were designed as perks to entice people to stay in the party. The
selection of national committee members he found "excruciating". Lou did not
mention how the SWP process worked, or what he thought might be an
appropriate alternative procedure. However, he was adamant that
"Zinovievist" type formations like the SWP could not accommodate real
leaders of the mass movement or insightful intellectuals in their

As I understand what Lou is saying, he feels the SWP process of leadership
selection was too restrictive, placing the bar too high in terms of
ideological homogeneity and agreement and not sufficiently low in terms of
openness to new leaders from the ranks. (Phil Ferguson seemed to feel, on
the contrary, that the process, as it operated in the New Zealand SAL, was
not selective enough; he complained that an NC slate included someone with
little experience or knowledge about Marxism.) These criticisms go to de
facto implementation, but not necessarily to the merits or demerits of the
procedure itself.

Of course any procedure, no matter how appropriately designed, can be
abused. But I think this "excruciating" national committee selection process
in the SWP bears some thoughtful examination. No better place to start than
with the theory, as it was first advanced by James P. Cannon in 1944. The
following exposition is taken from the SWP Internal Information Bulletin,
"Problems of Leadership Selection and Leadership Structure", April 1969. It
was originally published in Cannon's Letters from Prison.

The procedure was an American original, adapted (as Cannon's postcript
notes) to the particular circumstances of the SWP at that time. But much of
it is of general validity, and was applied for many years by the SWP and
other Fourth International sections and sympathizing groups. Of course there
are a number of aspects that are not explored in Cannon's exposition. For
example, more or less proportional representation of minority tendencies and
factions in leadership bodies is not mentioned, although this was common
practice, for example, in the Canadian and other FI sections.

As Cannon explains, the procedure he outlines is in sharp contrast to the
undemocratic slate procedures common in the social democracy and the early
U.S. CP. And it is profoundly democratic in conception, based on the
underlying principle that leading bodies must reflect, and be accountable
to, the rank and file membership of the party. Describing the importance of
the National Committee, Cannon says "control of the central leadership,
which in day-to-day practice is limited to a very small group, by a larger
group standing closer to the rank and file, is the most important mechanism
to assure the democratic half of the Leninist formula: democratic

It would be interesting to know what if any alternative methods of
leadership selection the critics have in mind.

   * * *


By Martin (James P. Cannon)

[Editor's note: Written from Sandstone Prison in 1944 and adopted as the
Nominating Commission procedure at the 1944 convention and all subsequent

In our opinion the most important reason for stretching the convention out
for another day is to give adequate time for a free and well deliberated
selection by the delegates of the new National Committee. This is one of the
strongest guarantees of the democracy of the party. Our party has always
been more democratic, ten times more democratic in this respect than any
other party. But there is room for improvement, and we should consciously
seek out the necessary methods. We never went in for any of the rigging,
wangling, vote-trading and leadership pressure devices by which, in all
other parties, the convention delegates are usually defrauded of a large
part of their democratic freedom of choice. If one has a self-sufficient
revolutionary party in mind, all such methods are self defeating. A
revolutionary party needs a leadership that really represents the party,
that is really one with the party. Without this democratic corrective,
freely brought into play at every convention, centralization and discipline
inevitably become caricatures and forms of abuse which injure the
organization every time they are exercised. A revolutionary leadership must
feel free at all times to act boldly and confidently in the name of the
party. For that, it needs to be sure that there is no flaw in its mandate.

No rules exist to guide us in the technical execution of this difficult and
delicate task to the best advantage of the party. The democratic selection
of the primary and secondary leaders is a sufficiently important question --
nobody knows how much damage can be done by bungling it -- but, as far as I
know, nobody has ever written anything about it. Nobody has taught us
anything. We are obliged to think and experiment for ourselves.

The democratic impulses of the rank and file incline them to react
unfavorably to "slates," as they feel, not without reason, that they narrow
down, for all practical purposes, the freedom of choice. The social
democratic politicians, who are as undemocratic a collection of rascals as
one can ever expect to meet, have always exploited this sentiment by
announcing their firm, democratic opposition to slates. Of course, there was
a little catch to their virtuous slogan of "no slates." They meant no
openly-avowed slates which would possibly be open to discussion and
amendment. Instead of that, the noble social democrats rig up secret slates
by means of horse-trades and petty bribes to ensure their control. A good
50% of social democratic convention "politics" is always devoted to this
kind of business.

>From the first days of American communism, which also coincided with the
first appearance on the scene of a new type of leaders with a new conception
of "politics," we tried to break through the "no slate" fraud and devise a
more honest system by which the leaders would take open responsibility for
their proposals and give reasons for their preferences in the make-up of the
leading committee. It became rather common practice for the leading
committees, in national as well as local conventions in the communist
movement, to propose a slate of candidates for the new committee to be
elected. We carried the practice with us in the independent movement of
Trotskyism. (During factional struggles the slate-making arrangements were
carried on in the separate caucuses of the factions.) This method was,
without doubt, far superior to the "no-slate" tricks of our socialist
predecessors, being more honest, and in the essence of the matter, even more

But this system, also, was not free from negative aspects, and even dangers.
I perceived some of them long ago, have thought much about the matter, and
from time to time have tried to devise corrective experiments. What
impressed me most of all was the quite obvious fact that while the
presentation of a slate of candidates by the leadership is the most
"efficient" way to get through the business of the election of the N.C. --
usually the last point on the agenda, carried through in a great hurry -- it
concentrates too much power in the leadership just at that very point -- the
convention -- where the democratic corrective of rank and file control
should be asserted most strongly. It is not the election of the central,
most prominent and influential leaders themselves. That problem solves
itself almost automatically in the interplay of party work and internal
strife. The problem arises over the selection of the secondary leaders, the
new committee members, the potential leaders of the future. As a rule, this
part of the slate, if presented by the most authoritative central leaders,
is accepted whether enthusiastically or not, by the convention; many
delegates are reluctant to oppose them.

It is senseless, of course, to speak of a revolutionary combat party without
recognising the necessity of a centralized, full-empowered leadership. But
this states only one-half of the problem. Leninist centralism is democratic
centralism, a profoundly dialectical concept. The other half of the Leninist
formula recognizes no less the necessity of subordinating the leadership,
really as well as formally, to the party; keeping it under the control of
the party. The party constitution does everything that can be done in a
formal sense to provide for the interaction of centralism and democracy. The
structure of the party is strictly hierarchical. Higher committees command
the lower. Full authority over all is vested in the National Committee. But
the N.C., like all other committees, is required to render accounts and
surrender its mandate at stated intervals to the party convention to which
it is subordinated. This is the formal, constitutional guarantee both for
centralization and the ultimate control of the leadership. But it is also
necessary to think about the spirit as well as the letter of the party
constitution. A far-sighted leadership should concern itself with the
elusive, intangible factors which can play such a great role in determining
the actual relationship between the N.C. and the ranks.

Some of these factors arise from the composition of the N.C. and the
division of functions within it. Nominally, this body consists of 25
members, and they all have equal rights. In addition there are 15
alternates. But the majority come to the center only for meetings of the
plenum, which are not held very often. Between plenums the power is
delegated to the Political Committee. From this it is quite clear that one
section of the National Committee is in a position to exert far more
influence on the day-to-day work and interpretation of party policy than the
other. Again, some are older, more experienced and more prominent than
others, and consequently, wield greater authority in the committee, as well
as in the party as a whole. On the other side, the committee members from
the districts and the younger members of the committee generally, who are
active in local work, are closer to the rank and file than the central
leaders of the party are, and represent them more directly and intimately.
This gives them a special function in the N.C. of extraordinary importance.
Their presence represents a form of continuing rank and file control and
supervision over the central leaders. They can fulfill this function,
however, only insofar as they are people of independent influence and
popularity in their own localities; only insofar as they are freely elected
on their own merits, not hand picked.

To be sure, the central leaders cannot be indifferent to the selection of
the secondary leadership. In this, as in everything else, leaders must lead.
In a certain sense, the central party leaders "select" their collaborators
and eventual successors. The question is, how to go about it? It is often
easy for politically experienced leaders to convince themselves that they
are better judges of the qualifications and potentialities of certain
candidates than the rank and file delegates. And, as a rule, it is not too
difficult to force their selections through by means of the "slate." This
may appear to be the most "efficient" way. But, in my opinion, there is a
better way.

Wisdom lies in "selecting" people who have popularity and influence in their
own right, and whose promotion coincides with the wishes of the party
members who know them best. That means, to select people who are advancing
under their own power.

I came to this conclusion a long time ago, and as far as I have been able to
influence the course of things, it has been the party method of selecting
the N.C. Extensive and varied experiences, with every imaginable kind of
experiment, has convinced me that this method, even at the cost of
incidental mistakes, works out best in the long run.

The central leaders of the party, who work from day to day without close
contact with the internal life of the branches, need such a constitution of
the N.C. if they are to lead the party confidently; lead it with the
assurance that they know the moods and sentiments of the ranks and are in
step with them. When doubt arises, or when some new important step is under
consideration, it is only necessary to consult the out-of-town members of
the N.C. by mail, or to call a plenum, in order to get a reliable sounding
of the party. Approval of a given course by the plenum is a pretty certain
forecast of similar action by the party. Conversely, when the plenum finds
it necessary to over-rule the Political Committee -- and this has happened
more than once, notably in 1938-39 -- it is a sign that the Political
Committee is out of line with the party and requires a change in its
composition. The 1938-39 National Committee rebuked the P.C. several times
and finally reorganized it, and later tests showed that the full plenum most
accurately reflected the sentiment of the party. A serious and conscientious
party leadership should deliberately aim at a National Committee so composed
as to be, in effect, a microcosm of the party. When the full plenum of such
a National Committee meets between conventions, to all intents and purposes
the party is there in the room. That is far more useful to responsible
political leaders than a roomful of handpicked supporters without
independent influence and authority. Bureaucrats who have special interests
of their own to defend against the rank and file need to surround themselves
with dependent henchmen, but revolutionary political leaders need support of
an entirely different kind, the support of people who really represent the
rank and file of the party.

There is another, and even more important, reason why the rank and file
convention delegates should take over the election of the National Committee
and be free from undue pressure and influence on the part of the national
political leadership in exercising this function. The free selection of the
full membership of the National Committee is perhaps the most decisive way
to strengthen and reinforce genuine party democracy. It puts the political
leaders under the direct supervision and control of a second line of leaders
who are in intimate daily contact with the local and district organizations
and, in fact, represent them in the plenum. This control does not have to be
exercised every day to be effective. The fact that it's there, and can be
demonstrated when necessary, is what counts. Strange to relate, the
professional democrats have never once in the history of our party bothered
their heads about the method of selecting the National Committee from the
standpoint of reinforcing party democracy. This, in my opinion, is because
they tend to think of democracy almost exclusively in terms of unlimited and
unrestricted self-expression, and forget that control of the central
leadership, which in day-to-day practice is limited to a very small group,
by a larger group standing closer to the rank and file, is the most
important mechanism to assure the democratic half of the Leninist formula:
democratic centralism.

Throwing the floor open for nominations on the last day of the convention is
not the only alternative to a slate presented by the outgoing N.C. That only
throws the delegate body into disorganized confusion and facilitates the
manipulation of the election by means of secret slates and horse trades, the
favorite method of social democratic pseudo-democrats.

There is no infallible formula, but the results of our experiments over a
period of many years argue most convincingly in favor of a slate prepared by
a nominating commission. Of course, there are nominating commissions and
nominating commissions. But the best, that is, the most democratic, is not
the nominating commission appointed by the outgoing N.C. nor the one elected
at random from the floor of the convention. The most efficient, for the
purposes set forth above, is the nominating commission selected by the
branch or district delegations on a roughly proportional basis -- each
delegation selecting its own representatives -- and then ratified by the
convention. The nominating commission, thus conceived, is a body actually
representing the rank and file delegations from the districts. It would be
grossly improper for individual central leaders to intrude themselves upon
the commission and seek to dominate its proceedings. That would amount to a
circumvention of the democratic process aimed at in the proposal. It is the
part of wisdom for the central leaders to leave the nominating commission to
its own devices, respecting the essence of party democracy as well as the

The nominating commission should be selected on the first day of the
convention; it should begin its sessions at once, and meet at least once a
day thereafter to consider the various nominations, until a slate is decided
upon for presentation to the convention when the election of the N.C. comes
up on the agenda. In my opinion, the first step of the commission at the
1944 convention should be to discard formally the ruling which paralyzed the
work of the nominating commission at the 1942 convention -- the utterly
stupid and reactionary principle that every member of the outgoing N.C. was,
as a matter of course, to be reelected unless good cause was shown to remove
him. That turns things upside down. Nobody can be "frozen" in any position
in a revolutionary party. He must stand for election at each convention and
the election must be free and open. Room must be left for competition and
rivalry and differences of opinion to operate without artificial restraints.
Members of the outgoing N.C. should be placed in exactly the same status as
new aspirants -- as candidates for election. The nominating commission
should adopt a rule to this effect at its first session.

The most practical next step is to take a preliminary poll to ascertain how
many candidates are generally favored for election as national leaders who
are not counted as representatives of any special district of the party.
This will clear the road for the apportionment of the remaining places on
the slate for local and district representatives. Here, again, there should
be no "freezing" of old representation and no automatic closing of the door
to new candidates from districts previously not represented. The object
should be to provide the fairest possible representation of the districts in
the new N.C. but the principle of proportional representation should be
modified by other considerations; the relative importance of the district;
the quality of the candidates; the special role played by certain
candidates, etc.

The commission should announce the time and place of its daily sessions, and
invite any delegate who wishes to argue for or against any candidate to
appear and take the floor. The slate finally decided upon, either by
agreement or majority vote, should be presented to the convention as the
nominations of the commission. That leaves a door open for other nominations
and free discussion before the ballot is taken. Naturally one would have to
have some good arguments for another candidate to hope to amend the slate of
the nominating commission. But if he thinks he has a strong case, there is
no reason why he shouldn't make the attempt. Adequate time and patience must
be accorded for the presentation of any such proposed amendments. The
heavens will not fall if a slate is amended once in a while.

One word more. The convention should not shunt the election of the new N.C.
off till the last hurried half-hour of the convention, when impatience of
departing delegations would tend to discourage full discussion and. ample
consideration of the various nominations. The best procedure would be to fix
a definite hour and day to take up the election of the N.C. whether the rest
of the agenda is finished or not at that time. This decision should be made
demonstratively in order to call sharp attention to the vital importance of
full and careful deliberation in selecting the party leadership. And even
more important, the convention will thus give itself time to do the job

All these measures will not guarantee the election of an ideal National
Committee. But they should help to provide us with the best Committee that a
free party can select from the material at hand by the method of party
democracy. If the returning delegates go home with the feeling that this has
been accomplished, the new N.C. will be able to begin its work with a strong
authority. On the other hand, the leadership, precisely because of the care
and deliberation taken in the selection of the personnel of the N.C., will
feel itself to be more than ever under the watchful supervision and control
of the party.

October, 1944

Note: If my remarks are sent abroad a postscript should be added to the
effect that they are designed for the method of selecting the N.C. in our
party, as it is today, at the given stage of its development, and are not
meant as a universal formula. Every party must work out its own methods on
the basis of its own experiences.

PLEASE clip all extraneous text before replying to a message.

More information about the Marxism mailing list