Gorgias, Not Plato Was the Archi-Stalinist
There are, roughly speaking, two philosophical approaches to an antagonistic constellation of either/or: either one opts for one pole against the other (Good against Evil, freedom against oppression, morality against hedonism, etc.), or one adopts a "deeper" attitude of emphasizing the complicity of the opposites, and of advocating a proper measure or the unity. Although Hegel's dialectic seems a version of the second approach (the "synthesis" of opposites), he opts for an unheard-of THIRD version: the way to resolve the deadlock is neither to engage oneself in fighting for the "good" side against the "bad" one, nor in trying to bring them together in a balanced "synthesis," but in opting for the BAD side of the initial either/or. Of course, this "choice of the worst" fails, but in this failure, it undermines the entire field of the alternative and thus enables us to overcome its terms. (Say, in politics, in the choice between organic unity and destructive terror, the only way to arrive at the truth is to begin with the "wrong" choice.) Therein resides the insurmountable difference between Hegel and the New Age notion of balancing the opposites.
Gorgias's On Nature, or the Non-Existent (the text survived only in summary form in Sextus Empiricus, and Aristotle's On Melissus, Xenophanes, and Gorgias) can be summed up in three propositions: (a) Nothing exists; (b) If anything existed, it could not be known; (c) If anything did exit, and could be known, it could not be communicated to others. If there ever was a clear case of the Freudian logic of the borrowed kettle (providing mutually exclusive reasons), this is it: (1) Nothing exists. (2) What exists, cannot be known. (3) What we know, cannot be communicated to others... But more interesting is the repeated "diagonal" mode of division of genre into species: Things exist or not. If they exist, they can be known or not. If they can be known, they can be communicated to others or not. - Surprisingly, we find the same progressive differentiation at the opposite end of the history of Western philosophy, in the XXth century sophistics called "dialectical materialism." In Stalin's On Dialectical and Historical Materialism, when the four features of dialectics are enumerated:
The principal features of the Marxist dialectical method are as follows:
Contrary to metaphysics, dialectics does not regard nature as an accidental agglomeration of things, of phenomena, unconnected with, isolated from, and independent of, each other, but as a connected and integral whole, in which things, phenomena are organically connected with, dependent on, and determined by, each other.
Contrary to metaphysics, dialectics holds that nature is not a state of rest and immobility, stagnation and immutability, but a state of continuous movement and change, of continuous renewal and development, where something is always arising and developing, and something always disintegrating and dying away.
Contrary to metaphysics, dialectics does not regard the process of development as a simple process of growth, where quantitative changes do not lead to qualitative changes, but as a development which passes from insignificant and imperceptible quantitative changes to open' fundamental changes' to qualitative changes; a development in which the qualitative changes occur not gradually, but rapidly and abruptly, taking the form of a leap from one state to another; they occur not accidentally but as the natural result of an accumulation of imperceptible and gradual quantitative changes.
Contrary to metaphysics, dialectics holds that internal contradictions are inherent in all things and phenomena of nature, for they all have their negative and positive sides, a past and a future, something dying away and something developing; and that the struggle between these opposites, the struggle between the old and the new, between that which is dying away and that which is being born, between that which is disappearing and that which is developing, constitutes the internal content of the process of development, the internal content of the transformation of quantitative changes into qualitative changes.
First, nature is not a conglomerate of dispersed phenomena, but a connected whole. Then, this Whole is not immobile, but in constant movement and change. Then, this change is not only a gradual quantitative drifting, but involves qualitative jumps and ruptures. Finally, this qualitative development is not a matter of harmonious deployment, but is propelled by the struggle of the opposites... The trick here is that we are effectively NOT dealing merely with the Platonic dieresis, gradual subdivision of a genus into species and then species into subspecies: the underlying premise is that this "diagonal" process of division is really vertical, i.e., that we are dealing with the different aspects of the SAME division. To put it in Stalinist jargon: an immobile Whole is not really a Whole, but just a conglomerate of elements; development which does not involve qualitative jumps is not really a development, but just an immobile stepping at the same place; a qualitative change which does not involve struggle of the opposites is not really a change, but just a quantitative monotonous movement... Or, to put it in more ominous terms: those who advocate qualitative change without struggle of the opposites REALLY oppose change and advocate the continuation of the same; those who advocate change without qualitative jumps REALLY oppose change and advocate immobility... the political aspect of this logic is clearly discernible: "those who advocate the transformation of capitalism into socialism without class struggle REALLY reject socialism and want capitalism to continue," etc.
There are two famous quips of Stalin which are both grounded in this logic. When Stalin answered the question "Which deviation is worse, the Rightist or the Leftist one?" by "They are both worse!", the underlying premise is that the Leftist deviation is REALLY ("objectively," as Stalinists liked to put it) not leftist at all, but a concealed Rightist one! When Stalin wrote, in a report on a party congress, that the delegates, with the majority of votes, unanimously approved the CC resolution, the underlying premise is, again, that there was really no minority within the party: those who voted against thereby excluded themselves from the party... In all these cases, the genus repeatedly overlaps (fully coincides) with one of its species. This is also what allows Stalin to read history retroactively, so that things "become clear" retroactively: it was not that Trotsky was first fighting for the revolution with Lenin and Stalin and then, at a certain stage, opted for a different strategy than the one advocated by Stalin; this last opposition (Trotsky/Stalin) "makes it clear" how, "objectively," Trotsky was against revolution all the time back.
We find the same procedure in the classificatory impasse the Stalinist ideologists and political activists faced in their struggle for collectivization in the years 1928-1933.  In their attempt to account for their effort to crush the peasants' resistance in "scientific" Marxist terms, they divided peasants into three categories (classes): the poor peasants (no land or minimal land, working for others), natural allies of the workers; the autonomous middle peasants, oscillating between the exploited and exploiters; the rich peasants, "kulaks" (employing other workers, lending them money or seeds, etc.), the exploiting "class enemy" which, as such, has to be "liquidated." However, in practice, this classification became more and more blurred and inoperative: in the generalized poverty, clear criteria no longer applied, and other two categories often joined kulaks in their resistance to forced collectivization. An additional category was thus introduced, that of a "subkulak," a peasant who, although, with regard to his economic situation, was to poor to be considered a kulak proper, nonetheless shared the kulak "counter-revolutionary" attitude. "Subkulak" was thus
a term without any real social content even by Stalinist standards, but merely rather unconvincingly masquerading as such. As was officially stated, 'by kulak we mean the carrier of certain political tendencies which are most frequently discernible in the subkulak, male and female.' By this means, any peasant whatever was liable to dekulakisation; and the subkulak notion was widely employed, enlarging the category of victims greatly beyond the official estimate of kulaks proper even at its most strained. 
No wonder that the official ideologists and economists finally renounced the very effort to provide an "objective" definition of kulak: "The grounds given in one Soviet comment are that 'the old attitudes of a kulak have almost disappeared, and the new ones do not lend themselves to recognition.'"  The art of identifying a kulak was thus no longer a matter of objective social analysis; it became the matter of a complex "hermeneutics of suspicion," of identifying one's "true political attitudes" hidden beneath deceiving public proclamations, so that Pravda had to concede that "even the best activists often cannot spot the kulak." 
What all this points towards is the dialectical mediation of the "subjective" and "objective" dimension: subkulak no longer designates an "objective" social category; it designates the point at which objective social analysis breaks down and subjective political attitude directly inscribes itself into the "objective" order - in Lacanese, subkulak is the point of subjectivization of the "objective" chain poor peasant - middle peasant - kulak. It is not an "objective" sub-category (or sub-division) of the class of kulaks, but simply the name for the kulak subjective political attitude - this accounts for the paradox that, although it appears as a subdivision of the class of kulaks, "subkulaks" is a species that overflows its own genus (that of kulaks), since subkulaks are also to be found among middle and even poor farmers. In short, subkulak names political division as such, the Enemy whose presence traverses the ENTIRE social body of peasants, which is why he can be found everywhere, in all three peasant classes. This brings us back to the procedure of Stalinist dieresis: subkulak names the excessive element that traverses all classes, the outgrowth which has to be eliminated.
In his polemics against Alain Badiou's reading of Saint Paul, Giorgio Agamben defines the singularity of the Christian position with regard to the opposition between Jews and Greek (pagans) not as a direct affirmation of an all-encompassing universality ("there are neither Jews nor Greeks"), but as an additional divide that cuts diagonally across the entire social body and AS SUCH suspends the lines of separation between social groups: a ("Christian") subdivision of each group is directly linked with a ("Christian") subdivision of all other groups. (The difference between Badiou and Agamben is that, for Badiou, this new "Christian" collective is the site of singular universality, the self-relating universality of naming, of subjective recognition in a name, while Agamben reject the title of universality.) The commonsense classificatory approach would tell us here: what's the big deal? Being Christian or non-Christian is simply another classification that cut across and overlaps with other classifications, like the fact that there are man and women which also cuts across all ethnic, religious and class divides... There is, however, a crucial difference here: for Paul, "Christian" does not designate yet another predicate (property or quality) of the individual, but a "performative" self-recognition grounded only in its own naming, i.e., a purely subjective feature - and, Badiou adds, only as such can it be truly universal. The opposition between objective-neutral universal approach and the subjective-partisan approach is false: only a radical subjective engagement can ground true universality. The constellation here is therefore exactly the same as the one with sub-kulaks in the Stalinist discourse: sub-kulaks are also the "remainder" of kulaks which cuts across the entire field, a subjective-political category masked as a social-objective quality.
So when Agamber defines "Christians" not as directly "non-Jews", but as "non-non-Jews",  this double negation does bring us back to the starting positive determination; it should rather be read as an example of what Kant called "infinite judgement" which, instead of negating a predicate, asserts a non-predicate: instead of saying that Christians aren't Jews, one should say that they are non-Jews, in the same sense that horror fiction talks about the "undead." The undead are alive WHILE DEAD, they are the living dead; in the same way, Christians are non-Jews WHILE REMAINING JEWS (at the level of their pre-evental positive social determination) - they are Jews who, as Paul put it, "died for (in the eyes of) the (Jewish) Law."
And, to go even further back (to Gorgias), one should read his argumentation in the same way. It may appear that Gorgias proceeds in three consequent divisions: first, things either exist or not; then, if they exist, they can be known or not; then, if they can be known, we can communicate this knowledge to others or not. However, the truth of this gradual subdivision is again the repetition of one and the same line of division: if we cannot communicate something to others, it means that we "really" do not know it ourselves; if we cannot know something, it means that it "really" does not exist in itself. There is a truth in this logic: as already Parmenides, Gorgias's teacher and reference, put it, thinking (knowing) is the same as being, and thinking (knowing) itself is rooted in language (communication) - "The limits of my language are the limit of my world."
The lesson of Hegel and Lacan is here that one should turn this dieresis around: we can only speak about things that DON'T exist (Bentham himself was on the trace of this in his theory of fictions) - or, more modestly and precisely, speech (presup)poses a lack/hole in the positive order of being. So not only we can think about non-existing things (which is why religion is consubstantial with "human nature," its eternal temptation); we can also talk without thinking - not only in the vulgar sense of just inconsistently babbling, but in the Freudian sense of "saying more than we intended," of producing a symptomatic slip of the tongue. So it is not that even if we know something, we cannot communicate it to others - we can communicate to others things we don't know (or, more precisely, to paraphrase Donald Rumsfeld, things we don't know we know, since, for Lacan, the unconscious as une bévue est un savoir qui ne se sait pas).
 Incidentally, Stalin's collectivization drive in the late 1920's (as well as two other "irrational" decisions with catastrophic consequences, Hitler's entry into war in 1939 and Slobodan Milosevic's expansionism in Yugoslavia of the early 1990's) demands a good old-fashioned economic approach. What drove Stalin to collectivization was not a demonic drive to perturb the return to normality in the (relatively) peaceful and prosperous mid-1920s, but the economic deadlock (the gap between the peasants and the meager industrial output). The same goes for Hitler: the German entry into war in the late 1930s was the only way to avoid impending economic catastrophy, so it was not that, after limited economic and politic successes in the mid 1930s, Hitler, instead of being satisfied with what he achieved, was driven into the catastrophic war by some irresistible demonic impetus. (And, mutatis mutandis, the same goes for Milosevic.)
 Robert Conquest, The Harvest of Sorrow, New York: Oxford University Press 1986, p. 119.
 Conquest, op.cit., p. 120.
 Giorgio Agamben, Le temps qui reste, Paris: Rivages-Payot, 2000. p.168.