29 May 2020

A Philosophy of Revolutionary Practice: The first two theses on Feuerbach [1977]

Jairus Banaji

The tract published below was written in India early in 1977, during the final months of the Emergency when theoretical work was a more or less enforced necessity, given that most practical activity had ceased. (A study that some of us did many years later showed that the massive strike wave of 1973–74 was flattened under the Emergency.) The essay was written as part of debates that had been ongoing for at least a year or more between comrades in Bombay, Delhi and Bangalore, and written as part of a transition in the life of their connected groupings from a strictly Leninist phase, when those of us in Bombay had called ourselves the “Revolutionary Bolshevik Circle”, to a phase when we began to refer ourselves as “the Platform Group” or “Platform Tendency”. The RBC had engaged in intense theoretical study and discussion, with the vague long-term aim of building a “party”. When this aim was finally given up, we had to look for more viable modes of political work, gravitating, inexorably, to the vast industrial areas of Bombay and Bangalore. Notions like the workers’ enquiry, “alternative” plans, and socially useful production became more meaningful than sectarian ideas of party-building, if only because they offered avenues to working with workers and unions that were more transparent and more immediately practical. One result of this phase of political reorientation was the Union Research Group (URG) which emerged from the end of the seventies and lasted for about ten years, after working with and helping to coordinate a very wide range of unions in the Bombay industrial areas. But this transition in perspectives never involved any attempt to move away from theory, on the contrary there was a very substantial production of theoretical work in areas like Marx’s crisis theory, the best Marxist understandings of fascism, domestic labour and the reproduction of capital, workers, unions and the more radical forms taken by the workers’ movement historically, Western Marxism, etc. As the essay itself shows, we read widely and wherever possible sought out literature in French and German which could be translated into English and discussed more widely. The essay itself attempted to argue that the relationship between theory and practice had to be rethought and mediated through some conception of how workers acquire a sense of themselves as making their own history and as bearers of a (still unfulfilled) historical mission, viz. that of emancipating humanity from the bondage of capitalism. 

May 2020


Theory is capable of gripping the masses as soon as it demonstrates ad hominem, and it demonstrates ad hominem as soon as it becomes radical. To be radical is to grasp things by the root. But for man the root is man himself.”


“I think that anyone’s political ideas … must be rooted ultimately in some concept of human nature and human needs. Now my own feeling is that the fundamental human capacity is the capacity and the need for creative self-expression.”


“Our differences are on subjects of some importance. The question at issue is whether the soul itself is entirely void, like a tablet where on nothing yet has been written … and everything marked on it comes solely from the senses and from experience, or whether the soul contains originally the principles of various notions and doctrines, which external objects simply recall from time to time, as is my view and that of Plato.”


“The truth of the hypothesis on the other hand lies in its perceiving that in the process of development the notion keeps to itself. It is this nature of the notion this manifestation of itself in its process as a development of its own self—which is chiefly in view with those who speak of innate ideas, or who, like Plato, describe all learning merely as recollection.”   


       “This Platonistic element in Humboldt’s thought is a pervasive one.”


       “We must look upon language not as a lifeless product but far more as a generative activity (eine Erzeugung)”

(von Humboldt)

       “The chief defect of all hitherto existing materialism—that of Feuerbach included—is that the thing, reality, sensuousness, are conceived only in the form of the object, or of contemplation, but not as human sensuous activity, practice (praxis), not subjectively.”


“The active creation of humanity has no other end than humanity itself. For humanity does not proceed outside itself while it is creating, nor does it produce anything new. Rather does it know that everything it creates by unfolding was already within it.”   

(Nicholas of Cusa)

“The practical creation of an objective world … is proof that man is a conscious species-being… Man produces even when he is free from physical need and truly produces only in freedom from such need… In the same way as estranged labour reduces spontaneous and free activity to a means, it makes man’s species-life a means of his physical existence.”


       “Practical philosophy, or, more accurately, the philosophy of praxis … this is the future fate of philosophy in general.”

 (von Cieszkowski)


Introducing the second edition of Ludwig Feuerbach and the End of Classical German Philosophy, Engels writes, in 1888, “In an old notebook of Marx’s I have found the eleven theses on Feuerbach… These are notes hurriedly scribbled for further elaboration, absolutely not intended for publication, but invaluable as the first document in which is deposited the brilliant germ of the new world outlook”.1 By “outlook” Engels meant, of course, what Lenin will later call the “hypothesis” of “materialism in history”. The function of the Theses is thus determined. They lay the basis for the constitution of Marxism as a “scientific sociology”.2This line of thought stretches from Lenin, through Adler, into Bukharin. Adler will put it very well. Marx realises the programme of Comte, or what Comte’s Positivism announces as a programme becomes, in Marx, a perfect science.3Despite its greater sophistication, Adler’s conception is substantially the same as the one Bukharin will propose later. Historical materialism, Bukharin says, is the general theory of society and of the laws of its evolution. It is sociology.4The Theses signal this turn. “Dissatisfied with philosophy, (Marx) throws himself into the hands of positive knowledge”.5

        The resistance that this conception will later encounter, in Gramsci, Lukács, Korsch, is too well known to recapitulate here. The conception of Marxism as sociology disintegrates the dialectic. And this line of thought stretches from the young Lukács into Sartre. Colletti will stand somewhere in between. There are “two aspects” in Marx, he says. “That of the scientist and that of the philosopher”.6And all this then sounds very much like the efforts of Fichte, Schelling and late 19th century neo-Kantianism to overcome the dualism that permeated, infected and destroyed Kant’s system.

There is another reflection of this same fundamental ambiguity. How did Marx’s ideas evolve? There is, first, the thesis of continuism. The theory of alienation is then an early prefiguration of the notion of abstract labour. In this development, Marx only abandons the speculative dross of his “early” writings. Or, second, there is a thesis of rupture. Marx in 1844 has still to establish his entry into the domain of Science. The Marx who has forced this entry cares nothing for his youthful mysticism. Obviously, Colletti and Althusser represent these alternative conceptions, with different degrees of emphasis.

But the “ideas” of Karl Marx evolved in a more subtle and complex way, as, indeed, most “ideas” do. In this essay, I do not propose to enter into any of these disputes directly. Instead I propose to concentrate on a certain phase in Marx’s development which can be dated, quite precisely, to the short period that intervenes between the summer of 1844 and the spring of 1845.

The concentration of interest on this period specifically arises in the following way. An earlier attempt to determine the thoroughly “rationalist” content of Marx’s method from the Grundrisse on led into a short critique of Colletti.7 This argued that Colletti simply restated Marx’s critique of Hegel in empiricist terms. The effort, made afterwards, to trace the roots of Colletti’s diluted and implicit empiricism led back to Della Volpe. In Della Volpe, a new dimension of “Western Marxism” became immediately apparent. The critique of Hegel proceeded there in a manner that was astonishingly close to the general positivist critique of “metaphysics”. How was such a “coincidence” possible? Was it a coincidence at all?

        And how to account for the fact that the only element of continuity in Marxists otherwise as opposed to each other on practically everything else, as Althusser, Timpanaro, and Colletti,8 was their common obsession with and hostility towards “Hegelianism”? By the sixties, there were no “Hegelian” Marxists left any longer. The Frankfurt School had disintegrated long before that, degenerating from its earlier “philosophical” interests into “critical sociology”. The anti-positivising opposition of this period would go back much more emphatically to Husserl and Phenomenology than to Hegel and Rationalism. So, how could one account for the regroupment of Althusser and Timpanaro, otherwise so opposed, on a common platform? The answer, it seems, lay in the deeper concern with materialism itself.

This raises the more general question of why a “philosophical” tradition or outlook is at all important to the Marxist movement. Why should it not proclaim pragmatism as its philosophy?

Thus initially, or methodologically, the purpose of looking for a “philosophical” legacy that Marx leaves to us is to find some basis in terms of which the esoteric history of the Marxist tradition becomes comprehensible. Precisely this, after all, was the task that Marx undertook vis-à-vis Hegel’s system, and as early as his doctoral dissertation.

Of course, a settling of accounts with the earlier generations of the Marxist movement is a vastly more complicated process, and one that presupposes other, more fundamental points of departure. But the abstract or philosophical moments of this critique are still essential. Take Marx himself. If, in The Holy Family, he traces the philosophical ancestry of the workers’ movement back to Locke’s empiricism, then this reflects, to him, not some arbitrary or whimsical presupposition. In Locke’sEssay, Marx sees a book that “systematized and theoretically substantiated the life-practice of that time”. Philosophy is seen here as a mediation between the “life-practice” of society in one epoch and the growth and development of the socialist movement in another.

This essay itself suggests no response to any of the questions raised above.

“Metaphysics will be defeated forever”

The pages in which Marx and Engels establish this philosophical genealogy for the socialist movement are among the most important they ever wrote.

In Chapter 6 of The Holy Family, they say,

“The French Enlightenment of the 18th century and in particular French materialism was … just as much anopen,clearly expressed struggle againstthe metaphysics of the 17th century, and against all metaphysics, in particular that of Descartes, Malebranche, Spinoza and Leibniz… Seventeenth centurymetaphysics, driven from the field by theFrench Enlightenment, notably, by French materialism of the eighteenthcentury, experienced avictorious and substantial [gehaltvolle] restoration in German philosophy, particularly in thespeculative German philosophy of the nineteenthcentury”.

Thus a “metaphysics”, and meaning here mainly the tradition of classical rationalism, driven into retreat by French materialism, restores itself in an even “weightier” form in the tradition that Hegel will finally bind together. Or, 17th century classical rationalism restores itself in 19th century German idealism. But,

“After Hegel linked it together in a masterly fashion with all subsequent metaphysics … the attack on theology again corresponded, as in the eighteenth century, to an attack onspeculative metaphysics andmetaphysics in general.” (All italics theirs)

That is to say, Feuerbach’s critique of Hegel has renovated the anti-metaphysical, or anti-rationalist, drive of the 18th century. And now (1844), given this historic renovation of the tasks of the Enlightenment, “metaphysics”

“will be defeated for ever by materialism, which has now been perfected by the work ofspeculation itself and coincides withhumanism”.

That is, Feuerbach’s materialism is not just the abstract mechanical materialism of the 18th century, it has absorbed elements from the tradition of “idealism” itself, and thereby “perfected” itself. This materialism, Feuerbach’s, is a humanism, and it will defeat “metaphysics” forever.

The paragraphs that follow deepen the conception of materialism itself by drawing out its historical ingredients in a more explicit way.

“There are two trends inFrench materialism; one traces its origin toDescartes, the other toLocke. The latter ismainly aFrench development and leads directly tosocialism”.9

The empiricism of Locke, “civilized” and thus modified by the materialism of the Enlightenment, forms the philosophical starting-point of the socialist movement. Or, the socialist movement traces its philosophical outlook back to the fusion of the “worldly” materialism of the French and the “common-sense” empiricism of the English.

Before we come back to this text, take a look at the central point that this “common-sense” empiricism makes in the person of Locke, at the very start of his “long awaited” Essay. Locke spends almost the whole of book one attacking the theory that human beings are born with “innate ideas”. More specifically, he wants to attack the notion that there are certain rational dispositions or capacities or structures that humans are born with, and that only these dispositions or capacities or structures can account for the disparity between their “knowledge” and their “experience”. Now, Locke himself can really only begin to tackle this thesis by first reducing it to the thesis that “therefore” in the production of “knowledge” experience plays no role at all. But finishing his polemic, Locke writes, “Let us then suppose the mind to be…white paper, void of all characters, without any ideas”, that is, without any rational dispositions or powers.

“How comes it to be furnished?… Whence has it all the materials of reason and knowledge?”

Or, if there are no rational dispositions of any sort given prior to experience, how is knowledge possible?

“To this I answer in one word, from Experience. In that all our knowledge is founded, and from that it ultimately derives itself… This great source of most of the ideas we have, depending wholly upon our senses, and derived by them to the understanding, I call “sensation”.10

Thus, Marx and Engels are saying that sensationalism, or Lockean empiricism, forms a basic ingredient of the socialist movement. As against which, Locke’s contemporary Leibniz, steeped in the “metaphysical” (that is, rationalist) tradition then dominant but later in retreat, presents the theory of “innate ideas” as most classical rationalists from Descartes on actually understood and argued it;

“The question at issue is whether the soul itself is entirely void, like a tablet whereon nothing has yet been written” or like Locke’s blank paper, or Holbach’s wax, “and everything marked on it comes solely from the senses and from experience, or whether the soul contains originally…principles…which external objects simply recall from time to time, as is my view and that of Plato”.

That of Plato, for he said, “all learning is merely a recollection” of what we already “know”. That is to say, experience is indeed indispensable in the production of knowledge, but it cannot by itself give rise to any knowledge. As Leibniz will say,

“The senses, although they are necessary for all our actual knowledge, are not sufficient to give us the whole of it, since the senses never give anything but instances, that is, particular or individual truths”.[11]11

Leibniz alludes here to something I shall come to in a moment. But let me repeat, or summarise in advance, the content of the argument that will follow. If it is indeed “humanism” that is to defeat “metaphysics”, then this must be a doctrine or a theory that makes some assumptions about human nature. Or, humanism will make no sense if it has no conception of what specifically makes human nature human and not simply natural. But, already implicit in the two rival conceptions of the “problem of knowledge”, that of Locke and that of Leibniz, are two entirely opposed conceptions of human nature. And one of these, the doctrine of Leibniz and classical rationalism, ascribes to humans certain rational dispositions, or powers, or faculties, or principles which, they say, can alone account for human intelligence or knowledge. According to this view, human nature is rational nature, nature that is specifically human by virtue of that element of “spontaneity” or “creativity” that we shall call reason. (Thus, throughout this essay, “reason” is understood as the set of all those dispositions and powers that the classical tradition called “innate ideas”.)

Now this precisely is not the view that either Marx or Engels accept, at least not in 1844. Rather, asThe Holy Family suggests in unmistakable terms, the conceptions of “knowledge” and of human nature in general developed there belong very much to the empiricist tradition. Later, I shall argue that theTheses on Feuerbach signal, however implicitly, a shift away from this position; that they propose a theory of knowledge in which the conception of human nature comes closer to the rationalist tradition in the same movement of thought that enables Marx, on the other hand, to supersede the epistemologies of both classical rationalism and classical empiricism. The notion of practice will dialecticise the opposition of “idealism” and “materialism”.

To return to Leibniz; in his reply to  Locke, he already alludes to the latent defect of all empiricist theories of knowledge. “The senses never give anything but instances”, he says. The inexorable logic of Lockean empiricism is Humean scepticism. From that point, it will bifurcate either into “personalist” and religious conceptions of “truth” (e.g. Kierkegaard and later existentialism) or into the conventionalist or pragmatist notions of “truth”. All empiricism, pursued consistently, must arrive at one or other of these three tendencies.

The Renaissance, despite its own philosophical nuances, shared an “ideal of rational certitude” in common. It shared in common the conception that our knowledge of the world can be, or ought to strive to be, a knowledge founded on certainty. Whether, with Hobbes, one espoused a geometricist notion of the procedures of rational thought, or, with Bacon, imparted to the process of knowledge a generalising or inductivist structure; whether with Leonardo you believed that experience is the “mother of all certitude”, or, with Galileo, one identified “real” being with “mathematical being”—knowledge itself was defined, normatively, by the goal of certitude. Thus, whatever their general philosophical inclinations, Aristotelian or Platonist, or their conceptions of the proper “method” of reasoning, they all “agreed that in any case the methods of modern science are such as to ensure that the conclusions it yields shall have about them nothing merely tentative, hypothetical or problematic”.12

But, with the triumphal march of empiricism initiated by Locke and Newton, this Renaissance ideal was quickly, and often only silently, abandoned.

The Renaissance itself had foreseen this inevitable tendency of all thorough-going empiricism. In the Aristotelian-empiricist notion of knowledge, Juan Luis Vives argued, “all universal principles are derived by us from particulars. Since these particulars are infinite in number, we cannot enumerate them all; but if one individual is lacking, the universal is not established”. And, from this, Vives concluded that there is therefore simply no solid foundation for a “demonstrative” science of nature.13

The problem foreshadowed by Vives or Leibniz bears the name of the problem of induction. It was a problem that led Bacon to rework the whole method of induction on what he thought were more sophisticated foundations, foundations that rejected immediate sense-observations and turned, instead, to experimentally-mediated observation as the real starting-point of knowledge. For the method of induction by simple enumeration produces

“conclusions that are precarious and exposed to peril from a contradictory instance; and it generally decides on too small number of facts and on those which are at hand”.14

It was a problem that led Hume to develop this underlying defect of empiricism into radical scepticism. For, “even after the observation of the frequent or constant conjunction of objects, we have no reason to draw any inference concerning any object beyond those of which we have had experience”.15 It was a problem which, in the form in which Hume left it, inspired Kant to write the Critique of Pure Reason, in order to secure solid epistemological foundations for our knowledge of the world. Hume “gave himself up entirely to Scepticism – a natural consequence, after having discovered, as he thought, that the faculty of cognition was not trustworthy”.

In his preface to the Prolegomena to any Future Metaphysics Kant wrote,

In order to do justice to the problem, however, the opponents of this celebrated man (Hume) would have had to penetrate very deeply into the nature of reason so far as it is occupied solely with pure thought, something that did not suit them”. 

“I openly confess my recollection of David Hume was the very thing which many years ago first interrupted my dogmatic slumber and gave my investigations… a new direction”.16

This “direction” would lead Kant to show, in his First Critique, that the very constitution of “experience” requires the intervention of “pure”, or a priori, principles of sensibility and the understanding. Finally, the same problem led to the disintegration and collapse of cognitive justificationism, of the view that knowledge is cognitively proven knowledge, and to the birth, early this century, of Poincaré’s conventionalism, Reichenbach’s probabilism, and Popper’s fallibilism, in short to the various modern bourgeois philosophies of science.


Sensationalism as an ingredient of Feuerbach’s ontological empiricism

By contrast, there is probably not a single passage in the whole corpus of his writing where Marx ever directly criticises the ideal of rational certitude, much less accepts a scepticist notion of our knowledge of the world. But, if Marx did endorse, to whatever degree, Locke’s conception of knowledge without ever consistently following through its epistemological implications, then this is so because he accepted empiricism, initially, not for its theory of knowledge at all, but at an entirely different level.

In the short history of philosophy that he and Engels present in The Holy Family, a heavy sense of contemporaneity dominates. In the attack launched by 18th century materialism on 17th century “metaphysics”, they see only a prefiguration of the attack that Feuerbach himself initiates in 1841 withDas Wesen des Christentums. It is Feuerbach’s image that stands in the background of their historical stage.

“Philosophy was counterposed to metaphysics, just as Feuerbach in his first resolute attack on Hegel, counterposedsober philosophy todrunken speculation”.


“But just as Feuerbach is the representative ofmaterialism that coincides withhumanism in thetheoretical domain”, and again, “But who revealed the mystery of the “system”? Feuerbach. Who annihilated the dialectic of concepts…?Feuerbach. Who substituted for the for the old lumber and for “infinite self-consciousness”…“Man”?Feuerbach, and onlyFeuerbach.”

That this fairly deep admiration for Feuerbach was not due simply to Engels or his collaboration with Marx is obvious if we glance, for a moment, at the Manuscripts that Marx had just finished writing before he worked with Engels onThe Holy Family. There he says,

“Feuerbach’s writings are the only works since Hegel… to contain a genuine theoretical revolution”.17

So, when, prophetically enough, as we shall see, they announced the final “defeat” of “metaphysics in general” at the hands of materialism, they announced the renovation, in Feuerbach, of the combined assault delivered much earlier by materialism on one side, empiricism on the other. They announced Feuerbach’s own programme:

“Philosophy has to begin not with itself, but with its antithesis, with non-philosophy. This is our internal essence, which is ‘unphilosophical’, absolutely anti-scholastic and distinct from thought. This is the principle of Sensualism”.18

To the abstract, semi-theological rationalism that Feuerbach sees in Hegel, he counterposes, in this principle of Sensualism, a sort of Lebensphilosophie that he callsUnphilosophie. This “unphilosophical” philosophy proceeds “from the truth and essentiality of the senses”. Reality, truth, require no mediation, says Feuerbach:

“Only that is true…which needs no proof, immediately speaks for itself and carries conviction”.


“Hegelian philosophy lacks immediate unity, immediate certainty, immediate truth”.

Beyond this sphere of immediacy, of the truth or the reality that is given to us in our experience of the world, in Sinnlichkeit,

“Man has no idea, no conception, of any other reality”.19

Marx and Engels announced the victory of this “materialism” prophetically enough. For what Feuerbach said was then said again in different ways, by Kierkegaard on one side, defending Subjectivity, Faith, Passion, Lived Experience, and by Mach on the other. What Marx and Engels witnessed early in the 1840s was only the first of a whole series of attacks on “abstract” rationalism that will gain in polemical intensity only later, towards the closing years of Marx’s own life.

Feuerbach’s critique of Hegel, to start with, was roughly contemporary with the efforts of Whewell and Mill to reargue the method of “eliminative induction”, or with Rankine’s own methodological anticipation of the later Positivist “philosophies” of science, and it preceded by barely twenty years the philosophical manifesto of positivism in Comte, or the fusion of positivism with natural science in Mach.20 If we then survey the whole 19th century from Herschel down to Poincaré we must definitely conclude that if the 17th century was a century of “metaphysics” and the 18th a century of “materialism”, thenthis 19th century was a century of empiricism.

And no longer the simple, uninvolved, dogmatic empiricism of the classical period, but a more complex, splintered, ramified empiricism, branching out from the narrow concerns of “epistemology” into fields left uncultivated by its classical predecessors.

Feuerbach himself is a case in point. Against “metaphysics in general” he opposes nothing so narrow or so circumscribed as a mere theory of knowledge (Locke, Hume), or a mere “methodology” (Whewell, Mill, Jevons) or a mere “philosophy of science” (Mach, Duhem). His opposition is more in the grand style of his own country, more “metaphysical”, so to speak. In sum, Feuerbach’s sensualism is not just empiricism pure and simple, it is an ontological empiricism, and it is this that accounts for the prima facie appearance that the concerns of Marx in 1843 and 1844 have little to do with empiricism as such, or that Marx’s thought had already moved into a sphere of its own.

It hardly needs to be established at length that Marx, like Hess and so many of the young Hegelians, was powerfully affected by Feuerbach’s sensualism. Marx’s early critique of Hegel was substantially only an elaboration ofmotifs that were central to Feuerbach’s own critique. This holds particularly for those themes or ideas in this critique that are today publicised (by Colletti and others) as specifically “Marxian”. Take the notion, brought back into focus by Della Volpe, that Hegel’s idealism depends crucially on a subject-predicate inversion and on the associated hypostatisation of the predicate. Already in theLectures on the Essence of Religion this motif is argued as follows,

“Man separates in thought the predicate from the substantive [i.e. from the subject in its traditional Aristotelian conception – JB], the property from the essence… And the metaphysical God is nothing but the compendium, the totality of the most general properties extracted from nature which, however, man … reconverts into an independent subject or being”.


“In religious ideas we have examples of how in general man converts this subjective into the objective, that is to say, he makes that which exists only in his thought, conception, imagination into something existing outside thought, conception, imagination”.21

Apart from this specific criticism, there is the general “humanism” that underlies it.

“The new philosophy makes human beings … into the sole, the universal, and the highest object of philosophy. It therefore makes anthropology, including physiology, into the universal science,”22

– a theme that runs through the whole of Marx’s Manuscripts.

Now I want to argue that it was precisely the ontological depth of Feuerbach’s empiricism that attracted Marx and Engels. In contrast to classical empiricism, Feuerbach was no longer mainly interested in a problem of knowledge; and in contrast to the earlier “natural-scientific” materialism, his ontology was one centred in a fundamental way on the relation of man to nature. It is true that the revived physiological materialism of Büchner and others drew a part of its inspiration from Feuerbach himself,23but Feuerbach would not accept this connection.24

Engels later describes the nature of the attraction that Feuerbach exerted on himself and Marx:

“Then came Feuerbach’s Essence of Christianity. With one blow it pulverized the contradiction [of Hegelianism – JB]… Nature exists independently of all philosophy. It is the foundation upon which we human beings, ourselves products of nature, have grown up. Nothing exists outside nature and man,” adding, “Enthusiasm was general;we all became at once Feuerbachians. How enthusiastically Marx greeted the new conception and how much … he was influenced by it, one may read inThe Holy Family”.25

But one could not simply absorb an ontology of the sort that Engels here describes, not even of the crudest sort represented by Büchner, Vogt and so on, without making some assumptions in the process about the “problem of knowledge”. So, it is worth looking a bit more closely at Feuerbach’s legacy.

Precisely because of its flaccid, ontological character, Feuerbach’s empiricism was far less precise, far more ambiguous or fluid, than the empiricism of Locke and the classical tradition. To start with, this fluidity was apparent in the very terms that were used to characterise the new “system” of philosophy. Feuerbach’s sensualism could just as well have been called his realism, or his realism his empiricism, or his empiricism his materialism, or his materialism his naturalism, or his naturalism his humanism. Whereas Marx accepted the Feuerbachian ontology mainly as a “consistent humanism or naturalism”, the later Marxist tradition, and this includes Engels, would see in Feuerbach mainly, or above all, a materialist.26

But, secondly, this fluidity only underlined another, deeper characteristic of Feuerbach’s system. His renovation of the 18th century attack pushed into the background precisely those themes that had distinguished or even defined classical empiricism. It reversed the traditional order of priorities of the classical empiricists. The latter, mainly concerned with the problem of knowledge, would leave the task of elaborating an ontology consistent with their empiricist epistemological positions to the much later tradition, in our own century, of logical atomism. By contrast, Feuerbach showed practically no concern with that problem. But definite epistemological assumptions were incorporated in Feuerbach’s sensualism. For,

“only thought which determines and rectifies itself by means of sense-perception is real, objective thought”.27

His critique of Kant likewise proceeds from the idea that the “objectivity” of thought, or “truth”, derives from our perceptions of the world:

“(for Kant) the objects of the senses are for the mind only phenomena, and not truth…What a contradiction, to sever truth from reality and reality from truth”.28

Expounding Feuerbach, and agreeing with him in this regard, Engels writes,

“the material, sensuously perceptible world to which we ourselves belong is the only reality”.

And Lenin,

“Feuerbach advocates objective sensationalism, i.e., materialism”.29

So, a philosophically implicit and unaggressive epistemological empiricism was one of the ingredients in Feuerbach’s many-sided legacy. And this is what accounts for the traces of an empiricist conception of knowledge in Marx, around 1844. It accounts for the fact that they are only traces, only implicit, sporadic, never consistently developed or reintegrated into the rest of his thought, and for the fact that they are traces of empiricism. So, in theManuscripts he writes,

Sense perception (see Feuerbach) must be the basis of all science. Only when science starts out from sense-perception … is itreal science”.30

And the Holy Family, following a month later, is more explicit. Here, Marx takes it for granted that

“Man draws all his knowledge, sensation etc., from the world of the senses and the experience gained in it”.31

Here, he praises Bacon as the “real progenitor of all modern experimental science”, for whom,

“Physics based upon sense-perception is the chiefest part of natural philosophy … According to him (sc. Bacon)”

But this is simply incorrect,32

“the senses are infallible and thesource of all knowledge. All science is based onexperience, and consists in subjecting the data furnished by the senses to arational method of investigation”.

This “rational method” is simply the method of induction itself.

“Induction, analysis, comparison, observation, experiment are the principal forms of such a rational method”.33

Here he criticises Hobbes for “systematising” Bacon’s materialism (which, of course, he never did),

“without however furnishing a proof of Bacon’s fundamental principle, the origin of all human knowledge and ideas from the world of sensation”.


“It was Locke who, in hisEssay on the Human Understanding, supplied this proof”.34

As if all this is not enough, Marx and Engels then go on to endorse a conception of human nature for which

“not only the soul, but the Senses too, not only the art of creating ideas, but also the art of sensuous perception, are matters of experience andhabit”.

And from this it follows, quite consistently, that

“the whole development of Man therefore depends on education andexternal circumstances”.35

One consequence of accepting such a view is the fact that when, in the Manuscripts, Marx attempts to define human nature, he will fail.

Before we come to that, however, we should take up the cycle inaugurated by Feuerbach, but at a phase closer to the closing years of Marx’s life-time. The explicit recovery of the merely underlying or implicit moment of ”epistemologism” in the renovated empiricism of the 19th century comes only with Mach.

Mach’s critique of “metaphysics”, the late 19th century inauguration of the positivist reaction that extends all the way down to the Vienna Circle, coincides in a remarkable way with Feuerbach’s. Both Feuerbach and Mach share the same conception of “metaphysics” as an illusion of knowledge, a deception of the human spirit, a thinly disguised form of theology.36

        More specifically, the notion of “hypostatisation”, central to the Feuerbachian critique of Hegel (both in Feuerbach and in Marx) recurs with the same critical function, and the same decisive importance, in Mach, and, elsewhere, in the tradition of American pragmatism, e.g., Dewey.37 It was this side of Mach that exerted the deepest or most lasting influence on the Vienna Circle. A logical positivist (Philipp Frank) who lived through the formative period of the Circle, during which it was in the process of absorbing ideas from Mach, Poincaré and others, writes,

“In my essay on the death of Ernst Mach, I characterized the enduring nucleus of Mach’s teachings as his struggle against the ‘idolization of auxiliary concepts’”.38

That is, Mach’s struggle against the “hypostatisation” of concepts that in our judgements have a purely predicative function (e.g., Feuerbach, Della Volpe) or a merely heuristic value (Mach himself). For Mach the limits of a valid cognitive discourse were circumscribed around the hard core of its “descriptive” terms, that is, those terms either directly describing, or reducible to descriptions of, our sensations. The residue of “metaphysical” terms such as “force” was strictly speaking eliminable or, if retained, then retained by some criterion of conventionalism. Later in his life Cassirer appears to have come around to this view. By the thirties, neo-Kantism had degenerated into a form of crude positivism.

“Cassirer repeatedly points out that science creates auxiliary concepts, such as force or atom, in order to be able to formulate conveniently the theories it has set up at a certain time, but that at later times these auxiliary concepts freeze into essences, ‘ontological concepts’”.

Here, in the notion of concepts “freezing” into essences or becoming “ontological”,

“there is hardly left any contradiction to Ernst Mach’s purely positivistic conception of science”.39

Of course, the only logically consistent programme that was ever advocated for the elimination of such terms (hypostatised concepts, ontological concepts, metaphysical terms, abstractions, etc.) was the one proposed by logical positivism, e.g., Neurath, Carnap, or later Craig.40 “Consistent” because at least in this less metaphysical tradition of empiricism an effort was made to install specific criteria of analyticity and of cognitive significance. For example, the programme of “elimination” was supported by the verificationist theory of meaning, though this itself, if pursued rigorously, would, as Wittgenstein realised,41 impart to all the most significant prepositions of any science a purely “hypothetical” value, for example, it would simply pulverise the notion of a “scientific” history, for what possible “primary experience” could one adduce to “verify” propositions in history, or, what possible “genuine propositions” could a science of history contain?

But Mach’s critique of “metaphysics” simply reiterated a conception that is there already at the heart of Feuerbach’s attack on “drunken speculation”. Certitude lies in “immediacy”, whether this is the immediacy of our “sensuous, physical existence” or the immediacy of our “sensations” as such. And characteristically, there is already in Feuerbach an element of conventionalism, for “love of convenience”, he says, is one of the reasons why “abstractions” that have no “objective validity and existence”42are none the less proposed or entertained.

In their own way, and Feuerbach less consciously than Mach, both rediscovered a deeply and consciously anti-rationalist theme ingrained into the very structure of empiricist thought. Berkeley, for example, had no quarrel with the notion that our knowledge involves “general ideas”. To him such ideas were in themselves particular, and thus reducible. For example, the concept of a “triangle in general” would involve no more than the use of some particular triangle (i.e., the figurate conception of it) to “denote indifferently” a whole group of triangles. Berkeley’s attack was directed, rather, against “abstract general ideas”’ such as “force”, “attraction”, “corpuscules”, “action”, ”reaction”, “space”, “time”, “extension”. Ideas that were not only “general”, but involved, to one degree or another, abstraction from all concrete contents.43 Or take Newton. His whole polemic against “hypotheses”, the one that forms the successive versions of Rule IV of the Regulae philosophandi is rooted in an identical prejudice. “What is not deduced from phenomena [induced from them – JB] is to be called a hypothesis; and hypotheses have no place in experimental philosophy [science – JB]”.44 Of course, this is not how Newtonian science would itself proceed, and the function of Newton’sRules is thus a different one, a doctrinal attempt to bolster the method of induction. (cf. Feyerabend in n. 44.)

But Berkeley, Newton, Mach, the logical positivists are explicit and straightforward. This explicitness is closely related to the fact that they all consciously draw out their own epistemological positions. Not so the sensationalist epistemology of Feuerbach, or of Marx in 1844. This is neither very explicit nor consistent.45

The later Marxist empiricism

But thirdly, again as a consequence of the Feuerbachian legacy, this same epistemological ambiguity, or shamefaced empiricism, permeates so much of the Marxist tradition of the following decades. It is this that accounts for the rather bizarre character of Lenin’s effort in Materialism and Empirio-Criticism.

This text, as we know, was directed polemically against Bogdanov and the Russian followers of Mach.46 Now Mach’s historical influence, e.g., his role in shaping the development of modern empiricism, was strictly tied in with the polemical thrust of his sensationalism, and not with the specific epistemological nuances that he gave to it. Mach survived into our own century as an inaugurator of the positivist reaction47 and not as a “Berkeleian idealist” or a phenomenalist. Mach’s later popularity was founded on his positivising conception of science, on the polemic that he built against “metaphysics”, on the solutions that he proposed to the confusions of contemporary mechanics, and not on the phenomenalist foundations of his empiricism.

But Materialism and Empiriocriticism was more interested in Mach’s phenomenalism. Lenin fought Mach for his “subjectivism”, or his “idealism”48 or his notion that our sensations are not “signs of things, but on the contrary, a thing is a thought-symbol for a compound-sensation of relative fixedness” (Mach).

This contradiction was, in many ways, an experimentum crucis of the philosophical strength built into Marxism. In Machism, Lenin, that is Marxism, confronted the one tendency that more than any other throughout that period, more even than Comte’s positivism, would exert a decisive influence on the trajectory of Europe’s philosophical thought over the next thirty years. A critique of Machism could only really proceed if it challenged Mach’s empiricism. But Lenin fought sensationalism with sensationalism. He fought “subjective” sensationalism with “objective” sensationalism. (These are his words.)

If, to Mach, objects were simply complexes of our sensations, or, more precisely, of our “sensory elements”, to Lenin these objects exist outside us in a reality that is given independently of our perceptions and we form a conception of them, or of reality, through the “mental pictures” or “images” that “arise exclusively from sensations”.49 Or, our ideas are “impressions”, in Locke’s terms, created in our minds by the impingement of the external world. Lenin says,

“Our sensation, our consciousness is only an image of the external world, and it is obvious that an image cannot exist without the thing imaged”.50

Nothing of the sort was “obvious”, for classical empiricism, proceeding from the same premises, drove itself, in different ways in Berkeley and in Hume, to radically different conclusions.51 At any rate, sense-perception is, or forms, or equips us with, a picture, or an “image” [Abbild] of the reality that exists outside us.

Now this conception of our ideas of the world having the form of “pictures” was fairly common later in the 19th century. It recurs, for example, in the philosophical views of both Hertz52 and Boltzmann53, Mach’s contemporaries. Mach’s sensationalism had led him to the view that “explanation is nothing but condensed description”, that the concepts and laws of science have a purely “summative” or “mnemonic” character and function. In terms that Bachelard will use later, Machism refuses to accept the notion of any sort of break between “sensory knowledge” and “scientific knowledge”. “If all the individual facts – all the individual phenomena knowledge of which we desire – were immediately accessible to us”, Mach writes, “a science would never have arisen”.54 Against which, Boltzmann argued as follows in one of his longer popular writings:

“General phenomenology seeks to describe every group of facts by enumeration and by an account of the natural history of all phenomena that belong to that area, without restriction as to means employed except that it renounces any uniform conception of nature, any mechanical explanation or other rational foundation. This latter view is characterized by Mach’s dictum that electricity is nothing but the sum of all experience that we have had in this field and still hope to have. [This view sets itself] the task of representing phenomena without going beyond experience … but I think this is an illusion.

No equation represents any processes with absolute accuracy, but always idealizes them, emphasizing common features and neglecting what is different and thus going beyond experience. That this is necessary if we are to have any ideas at all that allow us to predict something in the future, follows from the nature of the intellectual process itself, consisting as it does in adding something to experience and creating a mental picture that is not experience and therefore can represent many experiences. Only half of our experience is ever experience, as Goethe says. The more boldly one goes beyond experience, the more general the overview one can win.”55

Whereas Mach traced the welter of confusions that mechanics had got into by the 1870s to the tendency to transcend the limits of “experience”, Boltzmann, in the introductory portions of the first volume of his Lectures on the Principles of Mechanics, published about ten years before Lenin’s polemic, wrote,

“It is precisely the unclarities in the principles of mechanics that seem to me to derive from not starting at once with hypothetical mental pictures but trying to link up with experience from the outset”,56a criticism that is strikingly similar in content to that which Marx directs against Ricardo and his “inability to forget profits” when dealing with value.

So, if Lenin referred to our “ideas” about the world as “pictures” or “mental images”, he did so in a quite different sense to Boltzmann’s. Lenin’s “pictures” were much closer, in their epistemological meaning, to Locke’s “impressions” than to Boltzmann’s “pictures” or to Hertz’s “images” [Scheinbilder]. And Boltzmann is a good example to take, because Lenin himself cites him, approvingly, against the Machists.

Now, Lenin argues not only that our sensations are “images” of the external world, but that they “give us faithful images”. And he adds, “This is Materialism”.57 Or,

“Our sensations give us an objectively true image of the external world”.58

Lenin upholds the empiricist thesis in its strong form, in a form that goes beyond Bacon’s empiricism, for Bacon does not maintain, as Marx supposed him to, that “the senses are infallible”, but says, on the contrary, that

“by far the greatest impediment and aberration of the human understanding proceeds from the dullness, incompetence, and errors of the senses”.59

If it is not already obvious by now that, by “materialism”, Lenin understood a straightforward empiricism, then we must turn to other passages where the point is repeated even more clearly.

“The first premise of the theory of knowledge undoubtedly is that the sole source of our knowledge is sensation”.60

“All knowledge comes from experience, from sensation, from perception”.61

In these pages, Lenin will make one of his rare early references to Hegel, endorsing the latter’s conception of empiricism as the doctrine that “finds the truth in the external world”.62 Elsewhere, Lenin notes that the “doctrine that deduces all our ideas from the experience of the senses, reducing knowledge to sensations” is called sensationalism, and adds the further remark, from a “dictionary of philosophy”, that,

“Objective sensationalism is nothing but materialism”.63

In a less bland, or more involved, form the same espousal of empiricism runs through the later tradition of Marxist theory, closer to our time. The most important case of this is the Della Volpean school. It would be quite pointless to concentrate attention, in this respect, on Colletti, for the whole of Colletti’s critique of Hegel was forged by Della Volpe, much earlier, much more profoundly, with more sophistication, and with deeper insight. But before coming to Della Volpe, recall a central motif of Marxism and Hegel, the one that Colletti summarises in the view that,

“Reality, in fact, is that which is objective, and the objective, contrary to idealism, is precisely that which is external to and independent of thinking subjectivity”.64

The nucleus of this conception is discernible, almost word for word, in the positivising reactions unleashed by Mach himself. Abel Rey, a “neo-Positivist” who accepted Mach’s theory of knowledge, is quoted at length by Lenin in support of “materialism”. Lenin writes,65

“Let us take the basic concept, the concept of experience. Rey assures us that Mach’s subjectivist interpretation … is a sheer misunderstanding…

But how does Rey, who accuses only the fideists of distortion, but not Mach himself, correct this distortion? Listen. ‘Experience is by definition a knowledge of the object…Experience is that over which our mind has no command, that which our desires, our volitions, cannot control, that which is given and which is not of our own making. Experience is the object that faces the subject’”.

And what is this view, Lenin then asks, if not materialism itself?

“What penetrating genius Engels revealed when he dubbed the latest type of adherents of philosophical agnosticism and phenomenalism ‘shamefaced materialists’”.

That is to say, Machism is “shamefaced” materialism. For Mach is a sensationalist, that is, a materialist, but he distorts or covers up this materialism with phenomenalism.

“The positivist and ardent phenomenalist Rey is a superb specimen of this type. If experience is ‘knowledge of the object’, if ‘experience is the object that faces the subject’, if experience means that ‘something external exists and necessarily exists’, this obviously amounts to Materialism!…  We are told (by Rey) ‘The objective is that which is given from without, imposed by experience; it is that which is not of our making, but which is made independently of us…’”.

To repeat, Colletti says,

“the objective is that which is external to and independent of thinking subjectivity”, Rey says, “the objective is that which is given externally, which is made independently of us”.

The conception of “objectivity” that unites the Marxist Colletti with the positivist Rey is the one that Hegel would call, in the minor Logic, the “common sense” notion of objectivity:

“In the language of common life we mean by objective what exists outside of us and reaches us from without by means of sensation”.66

From Feuerbach’s sensualism, through the materialism of Lenin and the positivism of Rey and Colletti down to the naïve realism of common-sense,

“Reality … is conceived only in the form of the Object” (Thesis 1), that is, as “the object that faces the subject” or as “that which is external to and independent of subjectivity”.

Della Volpe is no longer concerned merely with these ontological conceptions of experience or objectivity. He recovers the classical interest in the problem of knowledge, and attempts to build an epistemology consistent with materialism. This entails a sort of synthesis of the decisive “critical” moments of earlier philosophical traditions. Specifically (i) of the various polemical refutations of the purely “negative” and even “romantic” conception of sensation characteristic of Platonism in general and of Hegel in particular, ranging from Aristotle’s critique of Plato, through Galileo’s critique of neo-Aristotelian apriorism and Kant’s refutation of Leibniz, down to Marx’s critique of Hegel;67(ii) of the “negative” or falsificatory inductivism of Bacon, purged of its hypotheticism, the defect, for Della Volpe, of all positivist conceptions of science; and (iii) of a diluted and modified Platonism from which Della Volpe derives his “theory of reason”.68

I do not want to summarise this long and involved argument any more than is strictly necessary for the purpose of this essay, so I shall confine myself to the empiricist legacy in Della Volpe’s reconstructed materialism. To Della Volpe, or in the conception of “scientific method” that he outlines, through the symbol of Galileo – but, remember, this is Della Volpe’s Galileo, and there are many other Galileos69 – the “instance” or principle or moment of Reason, represented by our “hypotheses” or “deductions” [ragione,ossia idea-ipotesi], can never establish its “truth” or its “validity”, that is, the “hypothesis” can never establish itself as “causal law”, simply by reference to itself, or tautologically, or without reference to the opposing instance or principle of “matter” as against reason, or of “facts” as against “hypotheses”, or of induction as against deduction. The validation of a hypothesis as law, i.e. the proof of its truth, if it bypasses or ignores this recourse to experience, fails to avoid or to overcome the logical fallacy of consequents, or the fallacy of believing that because, if the hypothesis is true, certain facts should follow, the hypothesis is indeed true because those facts are found. Thus, a hypothesis, representative of “reason”, acquires the proof of its truth, or becomes a “law”, by recourse to “experimental” observations of nature. But experiments, representative of the claims of experience, or of the “discreteness” that characterises reality as against thought (cf. Feuerbach)70 are useless in this regard unless they conform to the “inductivist and empiricist principle of Bacon comprised in the notion of an ‘excluding instance’ or a ‘negative instance’”71. Thus, for Della Volpe, experimental proofs proceed by disproofs. An experiment can only validate one hypothesised fact by “eliminating and refuting” other competing facts. Or, to relate this conception more closely to the bourgeois philosophies of science, Della Volpe retains a justificationist view of knowledge, but disposes of its abstract immediacy as a merely simple, direct, or unmediated intercourse with nature. He validates or endorses a type of justification that proceeds only “negatively”, as falsification, a type of proof that is solid only as disproof. For, “where a hundred instances will not prove a universal connection, one will disprove it. This is the cornerstone of  Bacon’s method: maior est vis instantiae negativae”. “Facts cannot prove an hypothesis by their agreement with it [for this commits the fallacy of affirming the consequent – JB] except insofar as at the same time they disprove its rivals by their disagreements”.72 It follows also that, to Della Volpe, all necessity is “factual necessity”, or necessity characterised by the contingency of the empirical order. As contingent necessity, a “law” is something intrinsically “corrigible” or “perfectible” with none of the fixity that characterises “metaphysical” thought.

        Della Volpe’s conception of science is thus strikingly close to Popper’s. Both are fundamentally concerned with what Popper himself calls the “demarcation problem”, the problem of establishing criteria in terms of which science, or its discourse, can distinguish itself from “metaphysics” and other discourses. Both see the hallmark of scientific method and of scientific “rationality” in the subordination of our “rational” constructions to the test of “experience”. Both accept that “experience” can play this role only “negatively”, or as falsification. Both regard the “theories” and “laws” of science as merely tentative, or conjectural, so that between this conception and those of the Vienna Circle or Popper’s hypothetico-deductive empiricism only a nuance intervenes.73

And, in both conceptions, there lurks the silent threat of positivism. Popper will rebut this charge by recasting positivism as a theory of meaning, as verificationism. Della Volpe will distance his conception from positivism by supposing that all positivism is characterised by its “idolatry of facts and its repugnance toward hypotheses”74. But all this might have carried conviction in the heyday of logical positivism with its anti-metaphysical “excesses”. Today, when this crusade has lost its original force and this classical, or rigorous, or pure, positivism has fragmented into the diluted, neoclassical, liberalised positivism of Hempel and others, both Popper and Della Volpe submerge in the general movement of modern empiricism, Popper as its liberal conscience, and its right-wing, Della Volpe as its metaphysical soul and its left.

The historical conjuncture that accounts for this strange, almost paradoxical convergence is something I shall not go into here. It is enough to note, for the moment, that both Popper and Della Volpe started their own speculations with Hume. With the breaks in justificationism that came towards the close of the 19th century, the problem of induction would again surface in an effort to reconstitute the legitimacy of empiricism.

Thus, Della Volpe could only secure the consistency of his epistemological construction, of the “functional reciprocity of reason and experience”, with the conception of “scientific method” proposed by him, and identified, symbolically, with Galileo, that is, with the “reciprocity of hypothesis and fact”, by subsuming, or accepting, or legitimising, a justificatory fallibilism, and thus presupposing the distinction between “fact” and “theory” that is common to all varieties of empiricism.

That such a distinction belonged centrally to the empiricist view of knowledge was already evident in the Marxist tradition itself when Lukács, for example wrote in an early essay, “The blinkered empiricist will, of course, deny that facts can only become facts within the framework of a system – he forgets that however simple an enumeration of ‘facts’ may be, it already implies an ‘interpretation’…”.75With this, Lukács anticipated a point that would become cardinal to the renewed disintegration of the contemporary empiricist philosophies of science.

        Della Volpe, however, required precisely this distinction of observational/non-observational statements because the notion of some sort of “empirical base” was crucial to the structure of his epistemological argument. Without this, which moment, within the field of cognitive discourse, could correspond to the ontological instance of Matter, or to the epistemological principle of “sensation”? Della Volpe could preserve the ontology-epistemology as a self-contained system, but without Bacon’s principle and its underlying notion of an “empirical base”, science and ontology would remain incoherently connected, and the very notion of a “scientific ontology”, that is, of materialism, would then collapse.76

Take a final and less important case of empiricism in Marxist writing. This is Schmidt’s The Concept of Nature in Marx.77In fact, Schmidt, when looking at Marx in 1844, will exaggerate his empiricism because of his failure to distinguish the sensationalist theory of knowledge from Feuerbach’s more wide-ranging, ontologising empiricism.

He says,

“In taking this view [of nature as the sole object of knowledge – JB] Marx showed himself to be rooted in the sensualism of Feuerbach and in fact he proceeded from sense perception as the ‘basis of all science’”.78

But Marx did nothing of the sort. He said that oneshould do so, and he referred to Feuerbach for authority. Schmidt is aware, naturally, of the central role that theTheses will occupy in the evolution of Marx’s thought. He is aware that the firstThesis makes what appears to be a concession to “idealism” (on this see below). Grasping instinctively, or intuitively, that this refers to the rationalist theory of knowledge, but committed, by virtue of an uncritical exposition, to the doctrinal assertion of empiricism in Marx, Schmidt compels his account to vacillate in a quite bizarre way between two opposed and conflicting conceptions of knowledge.

“Since men are forced to rely on material which exists independently of them, there is in fact nothing in their minds but what was previously present to their senses, as sensualist philosophers maintained”, 79that is, as the classical empiricists maintained.

Here Schmidt vindicates Locke against Leibniz. But, on the other hand,

“Perception itself … is based on conceptual operations”. 80

And this is more in the tradition of Kant, or of Leibniz, or of classical rationalism.

This confusion is then resolved by the argument, a strange one, that the rationalist, in this case Hegel’s,

“inversion of the sensualist principle that ‘there is nothing in the senses which was not previously in the mind’ becomes truer with the transition to the bourgeois era”. 81


“[u]nder preindustrial conditions the objective, natural moments are dominant, whilst in industrial society the moment of subjective intervention asserts itself in increasing measure… Marx also grasped the epistemological content of this economic fact, as can be seen from the first thesis on Feuerbach”. 82

This is one way of resolving the philosophical confusions one gets into when he/she starts, as Schmidt does, with the piously doctrinal notion that there is a clear logical continuity in Marx’s development. Schmidt is forced to reconcile deeply conflicting conceptions, conceptions that divide Locke and Leibniz, and ignores the fact that Marx would later reject Feuerbach’s empiricism and describe his own connection with it as “very humorous”. 83

Now, in all of the cases mentioned above (Lenin, Della Volpe, Schmidt, Colletti), it is a certain conception of materialism, one that Marx himself rejects in theTheses, that entails, or implies, the resumption of such empiricist positions. Hegel had already noted the internal connection when he called materialism of this stamp “Empiricism … systematically carried out”, that is, ontologically inflated and expanded into a conception of “reality” as a whole. In the Marxist tradition, this movement works itself out in reverse.

Sensualism and the rudiments of praxiology in the Manuscripts

As early as the Manuscripts Marx announces the programme of elaborating a “ consistent humanism” that

“differs both from idealism and from materialism and is at the same time their unifying truth”.84

        But it will take Marx roughly a year, the year that leads from the Manuscripts to theTheses, to draw out the real nucleus of this conception, the “germs of a new world outlook” (Engels). In theManuscripts, Feuerbach is totally exempt from criticism, and the argument proceeds from strictly sensualist premises. The consequence of this is that, in terms of the argument that is actually developed there, Marx can only end with a distinctly Kierkegaardian conception of man as a “suffering being”, who, “because he feels his suffering … is a passionate being”. And, as in Kierkegaard, so in Marx, this “passion” becomes the sign par excellence of “la realité du vécu’’, maintained, in both, against Hegel.85 

But a “passionate” or a “suffering” being is still only a natural being. This is the inner dilemma of theManuscripts. They stand halfway between two distinct definitions of what it means to be human.

The attempt here to overcome the false and abstract opposition of “spiritualism and objectivism’’ centres firmly on the relation of man to nature. Marx must overcome the false separation of subject and object without validating the “mysticism” of a “subjectivity that encroaches upon the object”, without validating Hegel, or the “false positivism” of his system, or the “formal and abstract” conception implied in the dialectic of self-consciousness. 86This, then, is Marx’s first effort to establish a real, and not merely formal, or abstract, or logical, tautoheterology.

For Marx will not simply say, ”man is man”, and he does not want to say only that “man is nature”.

“The true essentiality of any thing is not the predication of it as identical with itself, or as different, or merely positive, or merely negative, but as having its being in an other which, being its self-same, is its essence”.87

To establish the human essence as this real tautoheterological identity, and no longer as formal or abstract tautoheterology, i.e. of a “consciousness” that forms “the totality of its moments”,88 Marx roots the principle of identity in the conception of nature as sensuousness. “But nature, too, taken abstractly, for itself, and fixed in its separation from man, isnothing for man”,89 or it is “devoid of sense”90 or “abstract nature, nature as a thing of thought”.91Now, if we acknowledge that “sensuousness … is the essence of nature”, 92then it follows that nature is a human essence, and this “human essence of nature” is the “natural essence of man”.93Thus the conception of nature as sensuousness dialecticises the opposition of man and nature. But secondly, or more specifically, the conception of nature as “sensuous consciousness andsensuous need”94establishes “objectivity” as what it truly is, as a dialectic of intersubjectivity. This is not a simple reciprocal implication of subject and object, for objectivity is here constituted as a function of the determination of subjectivity as sensuousness. For, as a “natural”, “sensuous” being, man,

“is a suffering, conditioned and limited being, like animals and plants. That is to say, theobjects of his drives exist outside him asobjects independent of him; but these objects are objects of hisneed, essential objects”.95

In other words, to be “sensuous”, that is, “natural”, is “to have sensuous objects outside oneself, objects of one’s sense perception”, 96and, on the other hand,

“as soon as there are objects outside me…I am another, a realityother than the object outside me. For this third object I am therefore areality other than it, i.e.its object”.97

The identity of subject and object is thus established here not immediately, or abstractly, but through the specific conceptions of “objectivity” and “sensuousness” that Marx installs into the argument and through the dialectic that they operate.

But Marx remained thoroughly dissatisfied with this, for the mediated identity of subject and object accomplished by this argument still presupposed the abstract or immediate identity of man and nature. It presupposed a conception of man as a merely “natural, objective, sensuous being”, or a conception of human nature as merely sensuous nature, that is, nature. And that is why Marx will conclude these tortured and difficult paragraphs with the admission,

“But man is not only a natural being, he is a human natural being; i.e. he is a being for himself and hence aspecies-being, as which he must confirm and realize himself both in his being and in his knowing”.98

What distinguishes man as specifically human? human nature as human nature, and not just nature? Marx realises that sensuousness cannot solve this problem. For,

human sense, in its immediate and objective existence, (is not)human sensibility and human objectivity”.99

The clue to the answer that Marx will later elaborate lies already in the Manuscripts, but inconsistently. This answer is contained in the pages which say,

“The practical creation of an objective world, thefashioning of inorganic nature, is proof that man is a conscious species-being… It is true that animals also produce. They build nests, and dwellings, like the bee, the beaver, the ant, etc. But they produce only their own immediate needs … they produce one-sidedly, while man produces universally; they produce only when immediate physical need compels them to do so, while man produces even when he is free from physical need and truly produces only in freedom from such need… Such production is his active species-life”.100

Here is the conception that will drive Marx, after the Manuscripts and over the year that follows, to break with the sensualism of Feuerbach. For if “man produces even when he is free from physical need and truly produces only in freedom from such need”, then there is in human nature an element that stands apart from the rest of nature, an element that we can call “creativity”, but creativity here not in any instinctual sense, but as a function of the capacity to think, imagine, and reason. For there is one other passage in Marx’s work where he returns to this motif of what is specific about human production, a well-known passage written many years afterwards, and there he will say,


“But what distinguishes the worst architect from the best of bees is that the architect builds the cell in his mind before he constructs it in wax… Man not only effects a change of form in the materials of nature; he also realizes his own purpose in those materials. And this is a purpose he is conscious of”. 101

Precisely because the concern of these two passages is essentially the same, it is the contrast between them that is striking. This lies in the introduction, into the later passage, of the notion of “purpose”, i.e. of Reason, as we shall see shortly.

The Manuscripts are halfway, I said earlier. They are halfway between their own embryonic praxiology and the inherited sensualism of Feuerbach. And this is a tension that Marx fails to resolve in theManuscripts. For, though he is aware that historicity is a basic dimension ofhuman nature, and though he says, towards the end of theManuscripts, “We shall return to this later”, in theManuscripts he does not return to this, he goes on to discuss Hegel, and that is where they end.

But, in Theses, he returns to this. And these, precisely, are theses onFeuerbach. They signal the resolution of that conflict which accounts for the peculiarly difficult and involved character of theManuscripts.

Towards a conception of human nature

How, then, does Marx propose to overcome Feuerbach’s sensualism?

The First Thesis says,

“The chief defect of all hitherto existing materialism (that of Feuerbach included) is that … reality, sensuousness, is conceived only in the form of the object or of contemplation, but not as sensuous, human activity, practice, not subjectively”.

I want to argue that this thesis is formulated so much in Feuerbach’s own terms, the terms with which Marx will have little to do later, that its real content is easily lost. For example, even here, inTheses that are critical of Feuerbach, there is the same, implicit, but, in this case, only superficial, equation of “reality” with “sensuousness”. Thus, he writes, “reality, sensuousness…”, still very much in the style and manner of theManuscripts. Again, he says, “sensuous, human activity”, and this suggests that the real difference lies between “contemplation” on one side, and “activity” on the other.

Take Schmidt. His interpretation is very much the traditional one, and vitiated precisely by the shallow understanding of Marx’s meaning rendered almost inevitable by his (Marx’s) formulations.

“In Feuerbach, man the species-being, provided with merely natural qualities, confronts the dead objectivity of nature passively and intuitively rather thanactively and practically… Feuerbach’s man does not emerge as an independent productive force but remains bound to pre-human nature.Physical activity does, it is true, presuppose this natural basis… Allwork is work on a fixed being”.102

And so on. Here the “practice” that Marx opposes as a superior principle to Feuerbach’s sensualism is conceived as “physical activity”, “work”, etc. But to state the difference in this way is to leave enough room for Feuerbach to accommodate to the Theses. This is why Schmidt will eventually reject his own critical positions on Feuerbach and decide that

“the characterization given in Chapter One of Feuerbach’s role in Marx’s development would now be more positive. In a study of Feuerbach I have endeavoured to show that the very concept of “mediating practice” which Marx and Engels polemically turned against Feuerbach owes in fact a great deal to him”.103

But this rests on Schmidt’s own failure to grasp the content of Marx’s thesis. If one interprets the allusion to the “active side” developed by “idealism”,

        “Hence … the active side was developed … by Idealism” as the conception that “the world is mediated through the Subject”, as Schmidt does,104 this settles nothing, for the crux of the distinction lies precisely in the nature of this “mediation”. For “production” in the narrow and false sense of “physical activity” is still, after all, very much a process within nature itself (physical, from physis, ‘nature’). “Animals also produce”, Marx wrote, and then went on to describe how human production, or production as ahuman activity, is something deeper or less limited than mere “physical activity”.105

The real clues to the difference thus lie in the other terms:

“The chief defect of all hitherto existing materialism (that of Feuerbach included) is that … reality … is conceived only in the form of the object or of contemplation, but not as … human activity, practice,subjectively.”

That is to say, “reality”, which is “practice”, can only be understood “subjectively”, that is, in relation to a principle of subjectivity. Or, in the terms of my own argument, “practice” remains impenetrable if it is not grasped as implying, or asserting, a theory of human nature. To go back to the Manuscripts,history can only really become a principle in the argument if it is taken, or can be shown to be, or is understood as, a “confirmation and realization” of human nature, or of those “dispositions and capacities” and those “drives” that form man’s “species-powers”.[106]106Or, history, to be really, or truly, the “process of self-creation of man”[107]107, must imply a notion of reality as “the confirmation of man’s essential powers”. It must be shown to be “the reality of his own essential powers”.[108] 108

But none of this can actually be established if we reverse the terms of Marx’s argument and conceive “practice” as mere “sensuousness”, that is, like Schmidt, accentuate only, or mainly, the sensuous (i.e. physical) determinations of human activity. For sensuousness cannot establish why, or how, in what respect, human nature ishuman.

Here, Marx had simply no alternative. To break the vicious circle of the Manuscripts, he could only return to a conception of human nature that underlies not only the tradition of German “idealism” but practically the whole philosophical thought of the west. This refers, of course, to the definition, first proposed in this form by Aristotle, that

“Man is an animate being endowed with the power of reason”.

That is, to a conception of human nature as rational nature that recurs, closer to Marx’s own time, in Kant.[109]109 And this enables us to see that the notion of “practice” not only remains impenetrable if it is not grasped as a theory of human nature, butcannot be grasped as a theory ofknowledge. The praxiology of Marx is thus an epistemological humanism, a fusion of moments that lie separate and external to each other in the classical tradition of philosophy.

Thus Thesis 2 (to which I shall return later) says,

“Man must prove … the reality and power … of his thinking in practice”.

Thesis 8 that,

“All the mysteries which lead theory into mysticism find their rational solution in human practice and in thecomprehension of this practice”.

Thesis 1 will itself call practice, “practical-critical activity”.

And earlier, in the Manuscripts, where this line of thought is already prefigured,

“the history of industry and theobjective existence of industry as it has developed is theopen book of the essential powers of man, man’spsychology present in tangible form”.110

And, as we moveon into Capital, it is this line of thought that comes more sharply into focus, precisely in contexts that discuss production as a general human activity, and the specific character of production in bourgeois society.

Tautology and heterology in classical philosophy

The praxiology of Marx, sketched “hurriedly, for later elaboration”, still dense with the resonances of Marx’s fading Feuerbachianism, resumes the programme announced, but not accomplished, in the Manuscripts, viz. “consistent humanism”

“differs both from idealism and materialism and is at the same time their unifying truth”.111

First, idealism. This was a term by which Marx referred indiscriminately to the whole tradition of classical rationalism, which in “separating thought from sense-perception”112or in opposing “thought” to “sense perception” as a higher principle, validated the claims of Reason, or of the intrinsically rational determination of human nature, only “abstractly” or tautologically. One of the earliest and most definite expressions of this movement of thought is the Florentine Platonism associated with Marsilio Ficino and Giovanni Pico della Mirandola, a Platonism which, hostile to all empirical science, or, like Pico, consciously downgrading our perceptions of the world as intrinsically unstable and deceptive,113 fails to exert any decisive influence on the “New Science” of Galileo. Thus, if there is a Galilean Platonism, then this derives from the later, more rigorously mathematical and scientifically-oriented tradition represented by Tartaglia, reviver of Archimedes, and his pupil Ricci.

Dogmatism, or the tautology of reason, is the permanent temptation and latent tendency of the whole classical rationalist tradition. This rationalist tautology, grounded in the abstract separation of “thought” and “sense perception”, is in fact much less evident where it is generally sought, namely, in rationalist psychology and the thesis of “innate ideas” and must be found at a different level altogether. For the crude versions of the theory of “innate ideas”, the notion, in other words, that because innately humans are characterised by certain rational faculties, or powers, or dispositions, human “knowledge” is innate – this version is one that very few of the major rationalists will ever argue.114 Even Leibniz, a passionate proponent of this thesis, will agree that

“the external senses are necessary to us for thinking, and, if we had none, we could not think”.

Leibniz will then add, to clarify matters,

“But that which is necessary for something does not for all that constitute its essence. Air is necessary for life, but our life is something else than air. The senses furnish us the matter for reasoning, and we never have thoughts so abstract that something from the senses is not mingled therewith” 115

When we turn to passages of this sort, then the force of the critique that Kant later directs against Leibniz is undermined. When Kant says, “Sensibility is something quite positive and an indispensable adjunct to our intellect”,[116]116 this is something Leibniz would not have disagreed with.

That most rationalists were prepared to accept, even if only formally or abstractly, a tautoheterology of reason and experience, only shows that classical rationalism came much closer, despite its internal tendency, to a dialectical conception of knowledge than any of the numerous proponents of empiricism. But, at another level, the tautological character, or tendency, of classical rationalism becomes much more apparent. The methodologies proposed by it, identifying with the Renaissance ideal of rational certitude, in which Galileo saw the “equality” of the human intellect with the divine,117find in mathematics, and particularly in geometry, but here only under the influence of the scientific revolution itself (cf. Kepler, ubi materia,ibi geometria), thenormative model of all human cognition, including, as we know, of the “moral” or human sciences (Hobbes, Spinoza). It is this aspect of classical rationalism that Lukács defines as its central characteristic in some of the most brilliant pages written by any Marxist on such questions:

“The attempt to universalise rationalism necessarily issues in the demand for a system… The correct positing of a principle implies – at least in its general tendency – the positing of the whole system determined by it; the consequences are contained in the principle, they can be deduced from it, they are predictable and calculable… This notion of system makes it clear why pure and applied mathematics have constantly been held up as the methodological model and guide for modern philosophy”.

What Lukács calls the “problem of irrationality”, that is, of the “intractability of given contents to the concepts of the understanding”, thus expresses the same tendency of thought that Marx sees issuing in the “false positivism” of Hegel’s system.

“It is evident that the principle of  systematisation is not reconcilable with the recognition of any ‘facticity’ [Tatsachlichkeit], of a ‘content’ which in principle cannot be deduced from the principle of form and which, therefore, has simply to be accepted as actuality [Faktizitat]… Instead it must be wholly absorbed into the rational system of the concepts of the understanding… In this event, thought regresses to the level of naïve, dogmatic rationalism”.118

What Lukács here calls the “postulate” of systematic rationalism, that is, of the deducibility of all contents from a “basic principle”, expresses only the aggressive claims of a Reason that seeks to establish its power, or validate its truth, tautologically, or by reference to itself alone. But, even here, it is necessary to recognise the tensions, or the subtlety, of rationalist thought. Take Spinoza. Like so many of the rationalists, he identifies the form of mathematical or deductive reasoning as the sole method of attaining “true knowledge”. But Spinoza is then aware that the proof of a theorem is not what makes the theorem true, and he introduces, therefore, the further idea that “it belongs to the very nature of thought” that we form “true ideas”. But because these conceptions are never very clearly separated, the argument is permeated by ambiguity.

“It is clear…that certainty consists in nothing but the idea itself … certainty consists in adequate ideas. As truth needs no external sign, but it suffices to know the ideational essence of things (or what is the same, their adequate ideas) … it follows that the true method does not consist in looking for the signs of the truth of our ideas after we have acquired them, but in the methodical search for the truth, the adequate ideas”.

Thus, here the truth of our knowledge is something that requires no external procedure of verification, no reference to anything other than the idea itself. This whole process of self-validating thought begins not with “axioms”, for from these alone “the intellect cannot descend to the understanding of individual things, since axioms have reference to an infinity of things”, but with what Spinoza calls “definitions”:

“The concept or definition of a thing must be such as to enable us to deduce from it all the properties of that thing so far, of course, as it is considered by itself alone and not in conjunction with other things. This can be illustrated by our definition of the circle”.

But how can we arrive at such “definitions”? Because, he suggests, “it is quite clearly the nature of a thinking being to form true or adequate ideas”, or, “truth” is given by “the nature and power of the intellect alone”. 119

It should be obvious, finally, that Marx saw in Hegel’s dialectic of self-consciousness precisely a tautology of this sort, of a “consciousness” that establishes itself as the “object”, or “establishes the object as itself”, and which, in superseding this objectivity and alienation, remains “at home in its other-being as such”.

“Consciousness – knowing as knowing, thinking as thinking – claims to be the direct opposite of itself, claims to be the sensuous world, reality, life – thought overreaching itself in thought”.120

Marx is saying that Hegel’s dialectic is only formally, or abstractly, or falsely, tautoheterological. In fact, it is a tautology that leaves the moment of “otherness” – the sensuous world, reality, and so on – unvanquished and issues, therefore, in a “false positivism”. If you like, the “rational core” of Hegel’s dialectic, Marx argues, lies in its tautoheterological structure. Its “mysticism” consists in the fact that this is only “formally”, or “abstractly” a tautoheterology, for the “object”, or the moment of “otherness” [heteron] is only “objectified self-consciousness”, or the “movement” as a whole is only a movement of “consciousness”.

Thus, idealism develops the “active side”, or it asserts the principle of subjectivity, or of reason, but only “abstractly”, or formally, or tautologically.

Against all of which stand materialism, empiricism, naturalism as tautologies of matter or experience or nature. Here “sense perception”, or “sensuousness”, or “objectivity”, or “nature” are separated from “thought” and exhaust it. This dogmatising empiricism which “conceives reality only in the form of the object”, as pure givenness or facticity, or as a world “in itself”, or as “underivable, primordial being”,121 this dogmatising empiricism is likewise, symmetrically, but more crudely, without any of the subtlety of rationalist thought, characterised by the following conceptions. By the notion, firstly, that all our knowledge has its sole “foundation” in “experience”. In Lenin’s words, that our “Sensations give us an objectively true image of the external world”. And, just as the dogmatising rationalism of the 17th century was eventually compelled to displace the function of securing “truth” for our knowledge from the form of reasoning itself to “intuition”, or to what Leibniz calls “the light born within us”, or (cf. Hobbes)122 to “definitions” understood as nominalist conventions, so the conception that our sensations are, on their own, or with the aid of induction, sufficient to render a “true” account of nature, or of the world, would have to rely on a naturalistic doctrine of observation, the pure form of which is discernible in “Aristotelian” empiricism. For,

“[t]he demand that we base knowledge upon experience makes excellent sense in the Aristotelian philosophy where experience is defined as the sum-total of what is observed under normal circumstances … and what is then described in some ordinary idiom that is understood by all. Aristotelian empiricism, as a matter of fact, is the only empiricism that is both clear – one knows what kind of thing experience is supposed to be – andrational – one can give reasons why experience is stable…”123

Characterised, secondly, by a methodology the reverse of pure deductivism, the method of induction, the specific hallmark of Newton’s empiricism with its desperate and contrived efforts to secure this doctrine against the claims of Reason, called “hypotheses”; e.g., the attempts that characterise Newton’s successive drafts of his Fourth Rule of conduct of empiricism. The first draft says,

“In experimental philosophy [science – JB] one is not to argue from hypotheses against propositions drawn by induction from phenomena. For if arguments from hypotheses are admitted against inductions, then the arguments of inductions on which all experimental philosophy is founded could always be overthrown by contrary hypotheses. If a certain proposition drawn by induction is not yet sufficiently precise, it must be corrected not by hypotheses but by the phenomena of nature more fully and more accurately observed”.124

This rule rests, in short, on the pure fideism of experience, that blind faith, as mystical and irrational as all faith, in the certainty, or the accuracy, or the reliability, or the veracity, of our “sensations”. 125The intensely dogmatic form in which Newton casts his Rules only indicates the anxiety that runs through the whole tradition of empiricism, so that, between a “demonstrative” science of nature, or a science founded on certainty, and the method of induction, there is an absolute contradiction that can be resolved only in favour of one or the other. Thus, whereas classical, 17th century rationalism sought to secure the ideal of rational certitude by generalising the model of mathematical proof, classical, 18th century, empiricism would abandon this ideal forever, in the direction of agnosticism, or conjecturalism, or probabilism.126

Finally, when “systematically carried out,”127 empiricism will end with its own metaphysics,128 absorbing man into nature, or mind into matter. This is as true of Feuerbach’s sensualism or of Lenin’s materialism in 1908, as of the earlier “crude, simple, metaphysical materialism” that Lenin himself rejects (18th century French materialism, 19th century German materialism). In its “systematic” form empiricism thus vacillates between a conception of theidentity of man and nature (or mind and matter) asserted as an immediate,abstract identity, cf. Marx himself in theManuscripts, and aheterology of man and nature, affirmed in the notion that though man is a “part of” nature, nature remains a force greater than man, a force that conditions him externally, a situation which ‘’imposes itself” on him; cf. Schmidt, “In the Marxist dialectic … it is non-identity” that is, heterology or difference, “which is victorious in the last instance”;129or cf. Timpanaro, proponent par excellence,

“If materialism amounted merely to the recognition of a reality external to the subject, then Plato, St. Thomas and all their followers would also be materialists”.

A significant admission. But

“Materialism is not just ‘realism’… By materialism we understand above all acknowledgement of the priority of nature over ‘mind’, or, if you like, of the physical level over the biological level, and of the biological level over the socio-economic and cultural level”.130

But whichever form of this ontology one accepts, it remains true that reality is here “conceived only in the form of the object”, as an otherness, given, external and opposed to “consciousness”. In this materialism, the one that Marx attacks in Thesis l, empiricism thus reveals its true character as “a doctrine of bondage”.131

So, in all these forms, or at all these levels, a dogmatic empiricism, and there can be no other sort, confronts a dogmatising rationalism in an insoluble problematic. Against both these, Marx proposes, as their superior or “unifying” truth, the conception of practice as a theory of knowledge, and of this practice as the necessary form in which human nature “confirms and realizes” itself, or those powers that define its species.

It follows also that, if Marx’s praxiology is to break with, or to overcome, or leave behind, the tautologies of Reason and Matter, then practice must be understood as a tautoheterology, or it must be understood dialectically. The rest of this essay is, then, basically about this.

The tautoheterology of “practice”

(a) The Dynamic Rationalism of Entelechy

Writing a history, in this case, the history of philosophy, Hegel faces the problem of defining “development”.

“In order to comprehend what development is, what may be called two different states must be distinguished. The first is what is known as capacity, power, what I call ‘being-in-itself’ (dynamis). The second principle is that of being-for-itself, actuality (energeia)”.132

And these terms refer us back to Aristotle, in whose ontology all “being” [ousia] is conceived as “activity”. In the first place, “activity” in the Aristotelian-ontological sense, is simply the operation or functioning of “powers”, or capacities, or dispositions, or drives, that are in themselves merely innate, or implicit, or merely “potential”. And, conversely, “a thing can become only what it has the specific power to become, only what it alreadyis, in a sense,potentially”.133 But, secondly, for Aristotle, these “activities and functions are logically prior to powers”, that is, we can only understand “powers in terms of their operations. We understand the power of sight or vision in terms of the activity of seeing… We understand the power of thinking in terms of the activity of thinking”. 134

Or, in Marx’s terms, without “industry”, or production, or human activity in the broadest sense, we could not know man’s “psychology”. It would not be an “open” book, and therefore not be a book that we could read at all. The human essence would remain impenetrable.

Thirdly, in the Aristotelian ontology, “activities” divide into two classes, those directed to an external end, and those comprising their “end in themselves”. Activities of the latter sort are “entelechic”, or they have their “end” [telos] within [en] themselves.

So, when Hegel defines “development”, he defines it basically in the precise sense the Aristotelians gave to “entelechy”. A “development” is an “operation” or a functioning of “capacities” that lie implicit within the process and constitute their “goal”. Hegel illustrates this with an “example”, but an important one, which should hardly be called just an example.

“If we say, for example, that man is by nature rational, we would mean that he has reason only inherently or in embryo”.

To say that human nature is rational nature (as was said earlier) is to ascribe “reason” as an innate power or capacity to human beings. It is to say that “reason” is a species-power.

“In this sense reason, understanding, imagination, will are possessed from birth”.

Yet crucially,

“while the child only has capacities or the potentiality of reason, it is just the same as if this child had no reason”.

That is, a reason that is merely innate, or simply “potential”, is like no reason at all.

“Reason does not yet exist in the child since he cannot yet do anything rational”.

Or, being rational entails “doing” something rational. Or, a “power” that does not function or operate at all is as good as no power, for we would have no knowledge of it. To this principle, Aristotle’s, Hegel gives the name of “the principle of projection into existence”.

“The principle of this projection into existence is that the power cannot remain merely implicit, but is impelled towards development”.

Or, what is only “implicit” must become “explicit”, it must “operate” or it must develop.

So far, then, Hegel only repeats, or deepens, Aristotle’s conception. But now he adds a further determination, for,

“That which is implicit comes into existence… It certainly passes into change, yet it remains one and the same, for the whole process is dominated by it”.

So “development” is tautoheterology, an identity of identity and difference, of the one and the many, and so on.

To define “powers” or capacities or dispositions in this Aristotelian-Hegelian sense is to constitute those powers, etc., as generative principles. It amounts to defining all development as a “generative process”. And this permits an extension of the argument.

Marx says, all hitherto existing materialism conceives reality only “in the form of the object”. That is, as something finished, or dead, or external. Nine years earlier, in 1836, Wilhelm von Humboldt had proposed an identical criticism of the existing conceptions of language. He wrote,

“We must look upon language [Sprache] not as a lifeless product [wie ein todtes Erzeugtes] but far more as a generative activity [eine Erzeugung]”.135

It is of course from Humboldt, and from the whole rationalist psychology of the preceding epoch, that Chomsky will derive the notion of a “generative grammar”.136The theory of “generative grammar” resumes, or recapitulates, the Aristotelian distinction between “powers” and “operation” as the (linguistic) distinction between “competence” and “performance”. “Competence” here refers to those rules of “universal grammar” internalised in the speaker’s mind as innate properties of human intelligence. Without such an innate schematism it would be impossible to account precisely for the creative, and, of course, rationally creative, aspects of language-acquisition and use. For example, by a fairly early age, children are capable of uttering novel sentences merely on the basis of the finite and grammatically degenerate sample to which they have till then been exposed. And, yet, even a native language is in a very real sense something that is learned. It follows that, in this case, the “innateness” hypothesis does not exhaust, or dispose of, the necessity for a theory of “performance”, but specifies in advance what sort of process “learning”, or the acquisition of language, is. The innate cognitive mechanisms, or principles, or “ideas” are only set into operation, or only become active, with experience. Thus, in this case, one is dealing with a dialectic of “competence” and “performance”, or of “powers” and “activities”, or, more generally, of reason and experience. In short, dealing with tautoheterology.

This connection between the rationalist theory of human intelligence, or the Platonic theory of learning, and development as tautoheterology is not an extrinsic or accidental one, for,

“In the process of development the notion keeps to itself… It is this nature of the notion – this manifestation of itself in its process as a development of its own self – which is chiefly in view with those who speak of “innate ideas”, or who, like Plato, describe all learning merely as recollection.”137

(b) Intentions. Freedom as the Spontaneity of Reason

“The chief defect of all hitherto existing materialism … is that … reality is conceived only in the form of the object”.

That is, as an “objectivity” that is given, external to man, imposed on him, conditioning him. As the “objectivity” of postivising Marxism, or “of a world that stands opposed to us in its strangeness”. Hegel says this, and adds, it is a mistake to

“regard the antithesis of subjectivity and objectivity as an abstract and permanent one. The two are wholly dialectical”.[138] 138

By which he means,

“the object is not rigid and processless”.

Or, put another way, the “object” is defined by its “rigidity”, by being the merely immediate, or given, or external, only in that conception of “objectivity”, or “form”, as Hegel here calls it, proposed by the “mechanical” view of nature.139The chief defect of all hitherto existing materialism, Marx writes, is that

“reality is conceived only in the form of the object…not as human activity, practice”.

And Sartre, correctly,

“But if human reality is action, this means evidently that its determination to action is itself action. If we reject this principle, and if we admit that human reality can be determined to action by a prior state of the world or of itself, this amounts to putting a given at the beginning of the series. Then theseacts disappear as acts in order to give place to a series ofmovements”.140

That is to say, we would then be defining reality as human activity only to reconstruct this “activity” in the conceptions of “mechanism” or in the “mechanical mode of inquiry”. And this would be to regress to the level of behaviourism, chief opponent of a modern rationalist psychology. Moreover, “if the act is not pure motion, it must be defined by an intention”, Sartre argues, defining intention as “a choice of the end”. “The end, illuminating the world, is a state of the world to be obtained and not yet existing”.141In Hegel’s terms, in the section discussing teleology in the smaller Logic,

“In the End the notion has entered on free existence and has a being of its own, by means of the negation of immediate objectivity”.142

Sartre again,

“Since human reality is act, it can be conceived only as being at its core a rupture with the given”.143

Hegel again,

“Will looks upon the immediate and given present not as solid being, but as a mere semblance without reality.”144

Or, through their purposes or intentions or ends, humans bear an essentially polemical attitude to the facticity of immediate being, to the merely one-sided objectivity of a world conceived “mechanistically” as closed in upon itself. Subjectivity thus

“shows itself as a modifying and determining principle”145

Or, the “objective” world loses its initial appearance of pure “rigidity”, of the mere facticity of being, only in relation to man’s purposes, or his intentions, or his activity; or only by virtue of the principle of subjectivity determined as freedom.

This is the relation that Hegel will call the “teleological relation”,

“a syllogism in which the subjective end coalesces with the objectivity external to it… This unity is on one hand the purposive activity, on the other the Means, i.e. objectivity made directly subservient to purpose”.146

Thus, in labour, the relation of teleology par excellence,

“Man not only effects a change of form in the materials of nature; he also realizes his own purpose in those materials. And this is a purpose he is conscious of.”147

In fact,

“What has been said may also be expressed by saying that reason is purposive activity”.148

From the principle of the “projection into existence” it followed that a reason which does not “function” or “operate”, or which is not “active” (cf. Aristotle) is as good as no reason at all. In the intentional structure or purposive character of human activity the entelechy of man’s rationaldispositions or drives or powers becomes ateleology of man’s rationalpurposes. And, in this teleological relation, it follows also that the abstract tautoheterology of reason becomes the real tautoheterology of subject and object. Or, to go back for a moment, when Hegel said, there is no “abstract and permanent antithesis” between Subject and Object, that the two “are wholly dialectical”, then this is what he meant. For, in the realisation of their purposes, humans “transform themselves into theother of their subjectivity and objectify themselves, thus cancelling the distinction between the two”.149Thus, in the processes of labour and of human activity in general, man’s reason, or his rational dispositions and drives, mediates as his “purposes” or intentions or ends, remains the “power ruling these processes”, even if, as in all labour, only silently, or with “cunning”, or through the mediation of “objects”, so that in “objectifying” itself human reason, or man’s species-powers, or his rational drives have “only closed with themselves and retained themselves”.

Let me repeat some of the points that may be only implicit in this analysis.

1) “Teleology is the higher truth of mechanism”, 150or, in Marx’s words, “Nature, taken abstractly, for itself, and fixed in its separation from man, is nothing for man,” 151or “reality must be conceived as human activity, practice, subjectively”. All of which means the same.

2) This teleology, or the intentional-purposive structure of human activity must itself be a mediation of the entelechy of human-rational nature, for otherwise “freedom” will be only “freedom in form”, as Hegel says, or will only be “will in the form of contingency”, or caprice, so that freedom then becomes a pure contradiction, “to this extent that its form and content stand in antithesis”.152

3) An entelechy of reason not mediated, or determined, as teleology, or a reason not determined as conscious purpose, or as freedom, has no praxiological significance. It then remains, as in Aristotle, a “natural teleology”, a philosophy, or ontology, of nature.

And 4) the basic point, practice, to be really such, must be conceived as the tautoheterology of reason, or of man’s rational nature, or of his dispositions or drives or powers, manifesting themselves in activity, and thus engaging in their “development”. Only in this conception can practice and thus history become a “confirmation and realization” of man’s essential powers.

Thus it is Marx’s praxiology that draws out the profound significance of Hegel’s axiom that what is rational is real, and what is real is rational. For reality, conceived as human practice, now becomes the projection into existence of human reason, or of that which constitutes human nature ashuman nature. And, conversely, a reason that is not mediated tautoheterologically,as practice, is no reason at all. It remains something which we cannot know and thus “unreal”.

This is the conception (the one I have just tried to sketch, all too briefly) that establishes the “unifying truth” of idealism and materialism. Any other conception collapses us back into their insoluble antinomy. For we are then back with

(a) “Abstract thinking”, that is, a reason, or an intuition, or a creativity that finds no principle of confirmation, or development, or realisation, or a reason that is “valid” independently of its projection into existence, or its development, or practice. We are back with a principle that forms the latent tendency of all classical rationalism.

(b) “Objectivity”, or “sensuousness”, that is, a reality or a world that still lies opposed to us in its strangeness, a world closed in on itself as a merely “one-sided” objectivity, a world “imposed on us” by experience, a world that is a mere “object of contemplation”. And here we are back with the principle of all “hitherto existing” materialism, of sensualism, of naturalism, and so on.

Or, tautoheterology is essential. Reality must be conceived as “practice”, i.e. subjectively. Feuerbach does not see this,

“Feuerbach, not satisfied with abstract thinking, appeals to sensuous contemplation” (Thesis 5).

The philosophical and scientific ancestry of “praxiology”

The praxiological humanism of Marx was in fact “prepared” by a long history preceding the properly modern epoch of philosophical and scientific thought that began with Descartes. The hallmark of this modern epoch was the radical dualism of Cartesian ontology, that is, of a world divided and separated sharply into res cogitans andres extensa, into Mind and Matter, Soul and Body, Society and Nature. This dualism shattered the “conceptual unity of the world and opened up the possibility” of “extreme” solutions,153of the solutions we have already referred to. By contrast, and this contrast is very striking, both Greek thought and the thinking of the Renaissance strove towards the ideal of a unified conception of reality, and in this process, or effort, protracted over many centuries, and interrupted by revolutions in social life and productive regimes, anticipated, or was bound to anticipate, certain basic themes that Marx would later fuse into his notion of “practice”.

(a) Rational Nature, and Reason as Tautoheterology. The Greeks

Already in the pre-Socratic tradition, and specifically in Anaxagoras, two principles have been clearly established. There is, first, the principle that Galileo will later accept in the more profound conception that Plato and Renaissance Platonism gave to it, viz. that the world is intrinsically “rational”, or, that “reason (nous) is the origin of the world and of all order”.154 This principle Plato will develop later into his “Theory of Ideas”. There is, secondly, a corollary of this principle, namely, the notion that reason [logos] is likewise the “criterion of truth”, which Anaxagoras argues by attributing to our sensations an inherent deficiency, or an intrinsic fallibility.155 This says that deprived of “reason”, man is left with no means of perceiving the “rationality” of the world, that is, with no means of establishing “truth”. Or, deprived of reason, the senses are sterile, or not productive of truth. With the Sophists and here especially with Protagoras, this abstract, impersonal “reason” of Anaxagoras is argued “subjectively”. “Man is the measure of all things”, Protagoras is supposed to have said, “of that which is, that it is; of that which is not, that it is not”. About which Hegel writes, this assertion

“is in its real meaning a great truth, but at the same time it has a certain ambiguity, in that as man is the undetermined and many-sided, he may in his individual particularity, as this contingent man, be the measure, or else self-conscious reason in man, man in his rational nature and his universal substantiality, is the absolute measure”. 156

But Hegel accepts this principle of subjectivity to mean that “All content, everything objective, is only in relation to consciousness”,[157]157that is, “for man”. Thus, Protagoras expresses reason, or thought, as “real existence”,[158]158he understands it “subjectively”. Reason conceived subjectively is then further determined, by Socrates, as an ethical value, as “the Good”, which man “has to find from himself”. Or, Socrates says,

“Man has to find from himself both the end (aition) of his activity and the end of the world, and must attain to truth through himself”.

Truth is thus a “product mediated through thought”, that is, a function of human activity, for “thought” has been determined subjectively as “real existence”, so that Socrates establishes

“the infinitely important principle of leading back the truth of the objective to the thought of the subject”. 

Or, “all that has value to men … is contained in men themselves”.159 After which, Plato dialecticises the subjective and ethical principle of Socrates by ascribing to the “ends” or “values” that humans must discover the character of intrinsic determinations of human nature. Our discovery is thus only a recollection of what lies implicit within us. Or, “all learning is for us nothing other than a recollection (anamnesis)”, Plato writes,160 meaning thereby that “the process of learning is not such that something foreign enters into us, but such that our own essence becomes actualized”.161That is to say, the Platonic conception of “learning” [mathesis] prefigures the doctrine of “innate ideas”, but in its tautoheterological conception. To say that learning is a process of “recollection” is a way of saying that, without certain rational dispositions or capacities or powers, or without the “spontaneity” of human intelligence, we could have no knowledge. On the other hand, the process of gaining knowledge, or of “learning”, itself requires that these dispositions, or powers, or that this whole innate schematism of intelligence comes into play, or into “operation”, or function actively, through the stimulus of experience. So, “we must not think that the bald conception”, that is, the crude conception, which no major rationalist ever accepted,

“of innate ideas is here indicated. The idealism of Plato must not be thought of … as that false idealism … which maintains that … we do not learn anything, or are not influenced from outside, but that all conceptions are derived from out of the subject… Plato’s idealism is certainly far removed from anything of the sort”.162

That is, Plato does not accept a tautology of reason for which the moment of “otherness”, of experience, and so on, is of no significance. Rather, he “places the truth in that alone which is produced through thought, and yet thesource of knowledge is manifold – it lies in feelings, sensations, etc.”163This is a very important statement, for it tautoheterologizes the notion of “reason”, or it argues that our innate dispositions cannot develop without the stimulus of experience. Or, reason as the instance of unity mediates itself through experience as the instance of multiplicity, and it is through this process, dominated of course by its unity, that our knowledge is established or that we learn.

It was, furthermore, Plato who enunciated the abstract logic of this specific relation, who established, in other words, the dialectic as tautoheterological identity. In theSophist, he writes, about this, “the point of difficulty and what we ought to aim to show is that the Other is the Same, and that what is the Same is another”.164Early on in his own development, Marx will grasp this conception of dialectic as tautoheterology (to autontauton, the same; heteron, other);

“It is characteristic of the entire crudeness of ‘common sense’ that where it succeeds in seeing a distinction [an otherness, JB], it fails to see a unity [an identity, JB], and where it sees unity, it fails to see distinction. If common-sense establishes distinct determinations”, for example, if it distinguishes a faculty of “reason” from a faculty of “perception”, or if it opposes Reason and Sense-Perception, “they immediately petrify surreptitiously”, they become entirely opposed principles, points of departure for different systems of philosophy,  “and it is considered the most reprehensible sophistry to rub together these conceptual blocks in such a way that they catch fire”.165

Thus, the twin classical epistemologies of rationalism and empiricism both accepted as their common logical foundation the Aristotelian syllogistic transmitted through medieval scholasticism.

This logic of tautoheterology will then find its own specifically ontological principle in the Aristotelian conception of “entelechy”. The dialectic thus acquires a principle of motion. For “the Becoming of Herakleitos is a true and real determination, but change still lacks the determination of identity with itself”.166In “entelechy”, however, that which changes “passes over into change”, yet remains “identical with itself”. In “mere alteration, on the contrary, there is not yet involved the preservation of identity in change”,167that is, there is no tautoheterology.

Aristotle, moreover, validates an “ethics” for which “the Good is that which comprises its end within itself… It is that which, comprising its end within itself, is not desired for the sake of anything else, but only for its own sake”.168This, to Aristotle, is “happiness”, to Kant human life itself,169to Marx human freedom, for Marx writes that beyond the “realm of necessity,”

“begins that development of human energy which is an end in itself, the true realm of freedom”.170

And of course, between “happiness”, or “life”, or “freedom” in all these cases there is no difference whatsoever. For that “rational nature” which, to Kant, forms the principle of the “end in itself” is precisely that “human energy” whose development, unfettered by social conditions, is “happiness” itself.

Thus, in all these ways, the tradition of Greek philosophical thought establishes themes, or principles, or motifs, that lie embedded, however deeply, in communism, or in the praxiological humanism of Marx, which is its expression in theory. 

(b) Renaissance Experimental Rationalism

After a long break of several centuries, it is only in Italy and in the Renaissance, that so many of the characteristic themes and motifs of Greek thought will again surface and find proponents and antagonists. But, of course, what distinguishes the Renaissance from a mere revival of more ancient classical traditions is that its speculation takes place against the impending background of the modern Scientific Revolution. Thus, what distinguishes the writings of thinkers like Zabarella, Benedetti, Giordano Bruno, da Vinci, or Galileo himself is an interest, never very pronounced in Greek thought, with method, but this in the deepest philosophical sense, as an enquiry into conceptions of truth and of human knowledge in general.

In the Marxist tradition, the Renaissance exerts a peculiar fascination. First in Engels, and later, more obviously, in Gramsci. The specific quality of the Renaissance that attracts Engels is the sort of intellectuals it produces, thinkers whom Engels, despite his materialism, will call “giants”.

“It was the greatest progressive revolution that mankind had so far experienced, a time which called for giants and produced giants – giants in power of thought, passion and character, in universality and learning. The men who founded the modern rule of the bourgeoisie had anything but bourgeois limitations. On the contrary, the adventurous character of the time inspired them to a greater or lesser degree. There was hardly any man of importance then living who had not travelled extensively, who did not speak four or five languages, who did not shine in a number of fields. Leonardo de Vinci as not only a great painter but also a great mathematician, mechanician and engineer, to whom the most diverse branches of physics are indebted for important discoveries”.

And Engels adds,

“The heroes of that time were not yet in thrall to the division of labour, the restricting effects of which, with its production of one-sidedness, we so often notice in their successors. But what is especially characteristic of them is that they almost all live and pursue their activities in the midst of the contemporary movements, in the practical struggle… Hence the fullness and force of character that makes them complete men”.171

I have quoted this at length because the Renaissance conception of “Science” will become precisely an expression of this “completeness”, or “universality”, or many-sidedness. This not only in its anticipations of the enlightenment conception of science as something that must reach the “masses”, awaken people, stimulate them, render them conscious172but more specifically, or more narrowly, in the tradition that it inaugurates, which has no name, but which can best be called “experimental rationalism”.

The Renaissance, it is true, will, like later periods, oppose Reason and Experience. On its stage, Renaissance Platonism will confront and fight the empiricism of the Parisian Occamists and of the Aristotelians generally. The merits and demerits of “reason”, “mathematics”, “observation”, “experiment”, “experience” will be debated almost continually. But there is a big difference. In the thought of the Renaissance as a whole, there is more tension, more subtlety, or intricacy, and even more balance. In this respect, the figure of Galileo is indeed quite symbolic, but symbolic neither of medieval Platonism nor of modern empiricism, symbolic of the Renaissance itself, of all its richness and subtlety.

In this sense, the experimental rationalism of da Vinci or Galileo is an attempt, the first of its kind, to forge a “method”, or a conception of knowledge, that reflects the tautoheterology of reason, that peculiar and specific blend of “reason” and “experience” which is almost totally lost with the modern epoch and its sharp bifurcations into 17th-century rationalism, applauding “mathematics”, and 18th-century empiricism, turning to “experience”. For even those who, like Descartes, sought to reconcile the two principles, did so only externally, or mechanically, or “methodologically”, in early prefigurations of the induction-deduction circle.

Take da Vinci.173Superficially, and here quite unlike Descartes, for example, he seems to vacillate between an extreme empiricism and an extreme rationalism. Against apriorism, or the tautology of reason, he says, “Experience never goes wrong”; he says, “Experience is the mother of all certitude”. Thus, to da Vinci, a “knowledge that is born and that ends in the mind” cannot be scientific knowledge. Knowledge, he says, must be “born of experience” and must “terminate in definite experience”. For “all our knowledge has its origin in our perceptions”. “Therefore, ye observers, place no trust in those writers who with their imagination alone have tried to set themselves up as interpreters between nature and man”. But then, on the other side, against empiricism da Vinci will validate precisely the claims of “reason”. For, “no human science can be called true science unless it proceeds through mathematical demonstration”. And again, “There exists no certitude where some branch of the mathematical sciences cannot be applied”. How, then, to reconcile these conflicting demands? Da Vinci is a proponent, or theoretician, ofinstructed experience. His experience is not the unmediated observations of the world on the basis of which traditional empiricism seeks knowledge through generalisation. It is experience mediated, or instructed, or informed, by experiment. Thus,

“I will proceed by first making an experiment, because my intention is first to appeal to experience, and then by reason to demonstrate why such experience is constrained to work in such fashion. And this is the true role to be followed by the investigators of natural phenomena: while nature begins with the cause and ends in experience, we must follow a contrary procedure – that is, begin with experience and with that seek for the cause”.

Nature, da Vinci believes, is “infused with law”, that is, with rationality, but only reason, assisted by the senses, can discover and demonstrate this rationality. Thus “reason” and “experience” stand in “functional reciprocity” (so too Della Volpe), or da Vinci’s notion of “experience” is a complex one,174as this final quotation shows:

“Experience, the interpreter between resourceful nature and the human species, teaches us that those natural laws which work themselves out with necessity can operate only in the forms prescribed by reason”.

The case of Galileo is a more obvious and striking one. This Galileo, the proponent of experimental rationalism, is the Galileo of Geymonat, but also, to some extent, of Koyré. The tensions at the heart of these two modern interpretations only show that with Galileo, as with da Vinci, the balance between “reason” and “experience”, or the sort of relation that integrates them, is never clearly articulated, always brittle, or unstable, or ambivalent. For the experimental rationalism of the Renaissance is still a fluid doctrine, torn by opposing demands and interests, and struggling, within science, to unify theory and practice. Geymonat and Koyré represent, with different levels of emphasis, these contrasting claims.


“Aristotelian empiricism insists on ‘experiences’ that may serve as base and foundation to the theory. Galilean epistemology, aprioristic and experimental at the same time, offers in reply some experiments constructed according to a theory, experiments whose specific tasks are to confirm or invalidate the application to reality of laws deduced from principles that have their foundations elsewhere”.175

Thus Galilean science is “aprioristic and experimental” at the same time, Koyré says, but its experiments are a function of its apriorism. Elsewhere,

“The Aristotelian was perfectly right. It is impossible to furnish a mathematical deduction of quality. And well we know that Galileo, like Descartes somewhat later, was forced to drop the notion of quality, to declare it subjective, to ban it from the realm of nature. This at the same time implies that he was obliged to drop sense-perception as the source of knowledge and to proclaim that intellectual and even apriori knowledge is our sole means of apprehending the essence of the real”.176

To which Geymonat,

“Does it follow from this that experience plays a completely subsidiary role in the Galilean method?”

Geymonat’s Galileo is thus more balanced, but also more uncertain. From Benedetti, he derives “the mathematical element and the empirical element”, so that

“In Galileo there flourished on one side the influence of Archimedes; on the other side, was the spirit of the Renaissance technicians which made him eager to connect theory and practice inseparably, to link scientific explanation with empirical control”.

But the later Galileo, of the final treatise, is more emphatically Platonistic, so that Geymonat will eventually conclude that Galileo never reached a full understanding of the relation between reason and experience. Rather,

“He oscillated between recourse to the purest deductive method and appeal, no less energetically, to empirical observation”.[177]177

I shall return to this underlying ambivalence of Galilean science later. For the moment recall the argument proposed earlier. From this, it should follow that the rationalism that Koyré sees as the hallmark of Galilean science must, if it is to avoid the apriorism of a reason closed in on itself, and thus the very “paralogism” that Galileo attributes to the Aristotelian physics, functionally integrate the “experimentalism” in which Della Volpe, for his part, sees the sole determination of his method. The nature of this integration is what baffles Geymonat and Koyré. But one thing is certain. The “experience” integrated into the structure of Galilean science is no longer the gross, unmediated experience of later classical empiricism. It is “experience” mediated by the rational structures that make it possible. So Koyré claims,

“Observation and experience – in the meaning of brute, common-sense observation and experience – had a very small part in the edification of modern science”.

In this respect, Aristotelian science was more consistent or accorded better with “experience” than the science of Galileo. And, precisely because the Aristotelians held their “experience” against the results of Copernicus and Galileo,

“one could even say that they [sc. observation and experience] constituted the chief obstacles that modern science encountered on its way. It was not experience, butexperiment that nourished its growth and decided the struggle: the empiricism of the modern science is notexperiential, it isexperimental”.178

But an “experimental” empiricism is an empiricism no longer true to itself, or an empiricism forced to acknowledge the priority of reason or the constitution of “experience” by reason. Experiments are the labour of reason within science, the “translation into objectivity of the Notion which exists distinctly as Notion”. Recall that in Hegel’s conception of “teleology”, the coalescence of subjective purpose with the objectivity external to it finds its purest expression inthe means by which humans, in their activity, realise their purposes. Experimental techniques are such “materializations” of reason (cf. Bernard,179Bachelard), and it is striking how the very motif that we encounter in Hegel or in Marx as the “teleology of labour” (above) recurs now in Koyré:

“Far from being opposed to each other experiment and theory are bound together and mutually interdetermined… Indeed, an experiment being – as Galileo so beautifully has expressed it – a question put before nature, it is perfectly clear that the activity which results in the asking of this question is a function of the elaboration of the language in which it is formulated. Experimentation is a teleological process of which the goal is determined by theory”,180 that is, by reason, represented here as “theory”.

In short, the experimental rationalism of the Renaissance attempts to establish the tautoheterology of reason as the method of science itself. No longer, then, as an abstract epistemological conception, but as a principle valid, more specifically, for the rational cognition of nature or of the world. I want to turn now to the deeper characteristics of this conception that account for its intrinsic ambivalence.

The limits of cognitive reason. The dilemmas of experimentalism in cognitivist conceptions of “proof”

Unlike the purely geometricist or deductivist rationalism of the 17th century, Renaissance experimental rationalism attempts to preserve a “dialectic” of reason and experience. Geymonat writes that

“Galileo often affirmed in his letters, that mankind has two instruments for knowledge: ‘sensory experience’ and ‘rigorous demonstration’. Yet this does not mean that he saw an actual duality in these instruments or that he assumed some type of antithesis between reason and experience. On the contrary, he saw in them a profound dynamic unity”.181

To this “dynamic unity” I have given a more abstract or philosophical name. I have called it the “tautoheterology of reason”. The experimental rationalism of the early scientific revolution will attempt to tautoheterologise reason by virtue of its “experimentalism”. This effort is fraught with tension, or ambivalence, for a reason we shall come to.

For the moment, it is enough to note that, conversely, this Renaissance rationalism will validate a principle of experiment, or what I shall simply call “experimentalism”, as a function of its effort to preserve or elaborate adialectical content in rationalism. It is enough, for the purposes of my argument, if we see in this its sole contribution, or its only message.

That a principle of experimentalism underlies or is implicit in the Theses on Feuerbach is, of course, something well-known from the second thesis. The problem can then be stated as follows: how, or in what sense, is this principle to be understood? Whatever answer we accept, this answer must fulfil the following condition, a purely formal one; it must be compatible with, and preferably it must deepen, the conception of practice sketched earlier.

The Second Thesis is, in fact, Lenin’s favourite one. If the First Thesis is one that he hardly ever cites, then that is because its whole polemical thrust is directed to the sort of “materialism” that Plekhanov and Lenin would later endorse, in a pure reversion to Feuerbach. But the Second Thesis, by contrast, enables Lenin to establish a very basic principle, for him, of the “materialist” theory of knowledge, what he calls “the criterion of practice”. This can be stated, briefly, as follows; “practice” is the criterion of the “truth” of our knowledge of the world. Or, in his own words,

“the criterion of practice … for every one of us distinguishes illusion from reality”.

What sort of “reality” does “practice” confirm or validate? For example, practice “proves the correctness of the materialist theory of knowledge”, at the foundation of which, as Lenin tells us elsewhere in Materialism, stands the “naïve belief of mankind” in the existence of an external world, or “naïve realism”. So, in this case “practice” is a proof of common sense, or of our firm and unshakeable “belief” in the existence of a world outside us. In other words, in this case, “practice” refutes solipsism and all other varieties of “subjective idealism”. But more importantly, the same criterion validates the “truth” of ourknowledge of the world. It validates science. For example, it validates Marxism itself as the only scientific conception of history.182As he says,

“the criterion of practice, i.e. the course of development of all capitalist countries in the last few decades proves only the objective truth of Marx’s whole social and economic theory in general.”183

This is a line of thought, or a conception of “practice”, that Della Volpe will take up precisely in connection with the view that Marxism represents the “Galileism” of the social, or human, or moral, world. As we saw earlier, to Della Volpe, “experiments” represent the claims of “experience”, or “sensation”, or “matter”. It is the experimentalism of Galilean science, as Della Volpe understands this, that vindicates the “inductivism of Bacon”. Now to establish, or to vindicate, Marxism as this Galilean science itself, but applied to the social world, Della Volpe has to preserve a sort of symmetry between the science that he reconstructs through the symbol of Galileo and his understanding of Marxism. In fact, for Della Volpe there is only “one Science”, “one method” valid forall domains of reality, including Marxism itself. For example, if the experimentalism of Galilean science, as Della Volpe understands this, is its defining feature,184it must likewise be basic to Marxism. Here, of course, it is the “criterion of practice” that allows the comparison to be established. Recall, to start with, that in the field of cognitive reason, reason itself proposes, or establishes, only “hypotheses”. It is the function of the “experiment” to transform these “hypotheses” into “causal laws”. That is, to validate them as “true”. When Della Volpe turns to Marxism, he is therefore quite consistent, for his argument runs, the hypotheses of Marxism

“cannot be verified [non puo verificarsi], establish their reality or become laws [e diventare realtà-legge] except in and through that historical (and not just abstract) materiality [materialità] which is the specific feature of economic and socialpractice or experience… Thus the basic Marxian hypotheses of labour-value, surplus-value and so on … acquired thereality or truth [verità-realtà] oflaws when thepractical, economic and socialexperience of monopoly capitalism of the last fifty yearsconfirmed that a phenomenon as serious as the crisis, for example, could only be explained by the contradiction basic to capitalism … a contradictionpredicted in thehypothesised connection of labour-value and surplus-value”.185

Thus, for Della Volpe, the function of the “criterion of  practice” lies in “verifying” the hypotheses proposed by Marx and “transforming them into laws”.186Conversely, if “reason” is to avoid the “apriorism” or hypostases of metaphysics, it must be a hypothetical-predictive reason,187or its hypotheses must be “practically verifiable” [verificabile praticamente], that is, verifiable through “historical experiment” [attraverso l’esperimento storico].188 Or, to Della Volpe, social “practice” is “historical experiment”,189 and it is “experimental” by virtue of the criterion of practice itself, that is, as a function of validating, or “verifying”, or “proving”, our “hypotheses”.

Thus, the experimentalism of “practice” is, for Lenin and Della Volpe, mediated through the sphere of cognitive reason. Or, it is the hypotheticism of “reason”, of Della Volpean reason, that accounts for the experimental character and function of “practice”. And this hypotheticism is, in turn, crucial to Della Volpe if he is to secure a model of rationality (in Popper’s sense) that safeguards “scientific” thought against the abstract, apriorist dogmatism of “metaphysics”. As Popper would say, the rationality of “science” consists in its “honesty”, that is, in its willingness to submit its results to the test of “experience”, for without this (the demarcation criterion) what difference would there be between science and metaphysics? Della Volpe’s thought is identical in substance, only he is a Marxist and Popper had precisely Marxism in mind when he started on his search for the demarcation criterion.

In short, this is one way of preserving or establishing the principle of experimentalism. It is a straightforward empiricist way of doing so, for it is the hypotheticism of reason that secures the “experimentalism” of practice.

Of course, nothing could be further removed from Marx’s conception both of “science” on one side, and of “practice” on the other. Take the famous letter to Kugelmann, cited more often for what it says about the law of value than for what it says about scientific method.

“Even if there were no chapter on ‘value’ in my book, the analysis of the real relations which I give would contain the proof and demonstration of the real value relations. All that nonsense about the necessity of ‘proving’ the concept of value comes from complete ignorance both of the subject dealt withand of scientific method”.190

To repeat something I have said earlier, at the level of cognition, sense-perceptions are no criterion of truth. Or, “the sensible world as such is altogether void of truth”.191 Or, “Philosophy is written in this grand book, I mean the universe, which stands continually open to our gaze, but it cannot be understood unless one first learns to comprehend the language and interpret the characters in which it is written. It is written in the language of mathematics, and its characters are triangles, circles and other geometrical figures…”.192 Or again, “A scientific analysis of competition is not possible before we have a conception of the inner nature of capital, just as the apparent motions of the heavenly bodies are not intelligible to any but him who is acquainted with theirreal motions,motions which are not directly perceptible by the senses”.193 Or finally, to summarise all this with Anaxagoras, “reason (logos) is the criterion of Truth”.

Take Galileo, point of departure of Della Volpe’s effort to secure the “criterion of practice”. The system of physics that opposed Copernicus was a physics based on sense perception. This internally consistent system of thought argued, against Copernicus, that if the earth were indeed in motion, then its movement would entail a centrifugal force of such magnitude that all bodies not connected to the earth would fly away, and all bodies connected to it would lag behind. Against which, or against their constant appeals to the “observation” that heavy bodies falling from a height go perpendicularly to the surface of the earth, Galileo argues, these writers

“call upon their senses rather than their reason to clarify the effect… Just as I have never seen or ever expect to see the rock fall in any way but perpendicularly, just so do I believe that it appears to the eyes of everyone else. It is, therefore, better to put aside the appearance, on which we all agree, and to use the power of reason either to confirm its reality or to reveal its fallacy”.194

That is to say, our perceptions of the world may be either true or false, but only our “reason” can tell us which they are. It follows also that there are occasions when our reason must defy our perceptions.

“There is no limit to my astonishment when I reflect that Aristarchus and Copernicus were able to make reason so conquer the sense that, in defiance of the latter, the former became mistress of their belief”.195

All of which establishes a principle that we can call the principle of the cognitive priority of reason over sensation, or of the “force of abstraction” (Marx) over the power of perception. Now, it is precisely this principle which, when pursued systematically, or followed through consistently, accounts for the inherent ambivalence of every experimental rationalism, the tension embedded at the heart of its conception between the opposing claims of “reason” and “experience”. For, when the proponents of this view, the only valid conception ofcognitive reason, refer, with varying degrees of emphasis, to a recourse to “experience”, or to “experiment”, or to “verification”, that is, to the moment of otherness, then this is a tautoheterology of reason or a dialectic entirelyinternal to the field of cognition itself. That is to say, the experiment, or the recourse to “experience”, merely draws out that which lies implicit in reason itself. Or experiments are only the “materialisations” of cognitive reason. There is no domain of “experience” constituted independently of the interventions of reason.196 It follows, therefore, that a tautology of reason is the permanent temptation of all science properly conceived. That is to say, even experimental rationalism with its effort to balance the claims of “reason” and “experience” glides inevitably into a tautologising rationalism. For in the “experiment”, properly conceived, i.e. conceived rationally, the reason or the “idea” that is the source of the experimental enterprise, or the purpose of that enterprise, simply “remains at home in its other-being as such”. As Marx will say, but here with reference to Hegel,

“The abstract idea, which directly becomes intuition, is quite simply nothing more than abstract thought which relinquishes itself and decides to engage inintuiting”.197

And this, the reproach made against Hegel, is quite simply the innermost tendency of all cognitive reason; not an accidental or whimsical or arbitrary defect of such reason, but an expression of its intrinsic nature. For all science to be properly science, to be arational investigation into the world,must accept the principle of thecognitive priority of reason over perception. And if it accepts that principle, thenwithin the limits of cognitive reason the whole function of “experience” must necessarily become totally problematic. It is this problematicity of “experience”, the inevitable consequence of a consistently scientific outlook, that Galileo wrestles with. It is the intrinsic tautologism of science conceived rationally, and itcan be conceived in no other way, that accounts for the fluctuations, tensions, ambiguities of da Vinci, or of Galileo, and indeed of the Renaissance as a whole.

Thus, on the one hand, it is Galileo’s own ambivalence that really accounts for the conflicting and one-sided conceptions of his “method” or his epistemology proposed in more recent times, from Koyré on one side, to Mach and Heisenberg on the other. And, on the other hand, this ambivalence is no mere defect of Galileo’s own thought, no personal failing, as Geymonat seems to suggest, but built into the very structure of cognitive reason, an expression of the inexorable fate of all such reason.

        When Simplicio, the proponent of Aristotelian empiricism in Galileo’s Dialogue, asks Sagredo (Galileo himself), “Did you make an experiment?”, referring to the “illustration” of the ball falling from the top of the mast of a moving ship, Galileo declares, “No. And I do not need it, as without any experience I can affirm that it is so because it cannot be otherwise”.198 And this reply is not the reply of dogmatism, it is the reply of science. Not the voice of apriorism, but the voice of reason, apriorist by its very nature. Take another example. Sagredo asserts that, to understand mathematically the cause of an event “far outweighs the mere data obtained from the testimony of others, or even from repeated experiments”, to which Salviati, objective and impartial between Simplicio and Sagredo, then adds, “You speak very truly; knowledge of a single effect acquired by understanding its causes” prepares the mind to understand other facts “without need of recourse to experience.”199 And that is something da Vinci had already said, almost in so many words. “There is no result in nature without a cause; understand the cause and you will have no need of experience”.200

        Now, it is only the modern bourgeois philosophy of science of the last forty years that has begun, however hesitantly, to understand this. By this I do not mean of course the positive affirmation of the tautologising tendencies of all science that defines the work of Bachelard, but the more orthodox, or acceptable, tendency of Popperism itself. Justificationism, or the ideal of a cognitively proven knowledge, or knowledge as proven knowledge, loses ground and disintegrates rapidly with the increasing confusions of classical mechanics and the subsequent rise of modern physics.201 The latent weakness of cognitive justificationism has already been alluded to. As far as its rationalist version went, it was clear, or soon became so, that mathematical proofs are in themselves no indication of “truth”. Strictly logical deductions enable us only to make inferences, but not to “prove” in the sense of securing truth to a proposition. Rationalist versions of justificationism would thus displace the burden of securing “truth” either to innate properties of the mind (called “intuition”) through which certain propositions, called “primary truths” by Leibniz,202 or “definitions” by Spinoza, can be established as certain foundations of the process of cognition, or to some form of nominalism, as with Hobbes, where cognition rests on “principles” that “cannot be demonstrated”. For its part, justificatory empiricism would attempt to resolve the problem of a purely cognitive “proof” through the doctrine, pure assumption, that all our knowledge proceeds against the inspectable background of “facts”, of an “empirical base” or an accessible, that is, neutral, “primary experience”. Thus, justificatory empiricism would have to posit some such principle as the “autonomy of facts”,203 in terms of which “protocol sentences”, or a purely observational language, become possible. But against this empiricism, again take the example of Galileo.

“Galileo claimed that he could ‘observe’ mountains on the moon and spots on the sun and that these ‘observations’ refuted the time-honoured theory that celestial bodies are faultless crystal balls. But his ‘observations’ were not ‘observational’ in the sense of being observed by the – unaided—senses… It was not Galileo’s – pure, untheoretical – observations that confronted Aristoteliantheory but rather Galileo’s ‘observations’in the light of his optical theory that confronted the Aristotelians’ ‘observations’ in the light of their theory of the heavens”.204

The standard assumption of all empiricist doctrine that our “ideas” can be regulated by reference to an “empirical base” that is neutral or invariant with respect to “theories” has thus formed the key point in the disintegration of modern empiricism. Whether, with Feyerabend, one says that all “experience” is “mutable” or contains “natural interpretations” that are challenged and rejected by successive “theories”, or, with Kuhn, one says that our “perceptions” are paradigm-dependent, or, with Bachelard, that the “reality” which science deals with already “bears the sign of reason”, or, with Popper, that there is no purely “observational language”, or, with Lakatos, that our ‘facts” are constituted by the “interpretative theories” that we choose – the point remains fundamentally the same, it is the point Bernard made long ago when he said that “experience is the privilege of reason”.

The collapse of cognitive justificationism was then succeeded precisely by the notion that all our knowledge is purely “conjectural”, and subject only to refutations. The most sophisticated version of this latter-day hypotheticism is, of course, Popper’s. But Popper, denying the distinction between observational and non-observational statements, could only secure his refutations, and thus the whole doctrine of fallibilism, on conventionalist assumptions. Thus, Popper

“realizes that in the ‘experimental techniques’ of the scientist fallible theories are involved, in the ‘light’ of which he interprets the facts. In spite of this he ‘applies’ these theories, he regards them in the given context not as theories under test but as unproblematic background knowledge … he may call these theories ‘observational’”, that is, he may “decide” to treat them as if they really were “observations” of the world, “but this is only a manner of speech…”.205

It was this innocuous-looking element of conventionalism that would enable Feyerabend to break out of Popper’s camp, and Lakatos himself to adopt a much more radical conventionalism that sees the “hard core” of any scientific programme as protected from “refutations” by pure decision.

To summarise the argument; is there an instance or a principle or a moment of “experimentalism” in the notion of practice? Or, is there an “experimental function” of practice? One solution, proposed by Lenin, followed by Della Volpe, is called “the criterion of practice”. Practice tests or validates the truth of the contents of our cognition. Or, the experimental function of “practice” is here secured in the standard empiricist way, through the hypotheticism of reason. Lenin and Della Volpe simply recast “practice” in the role traditionally delegated to “experience” by classical empiricism. Experimentalism, here, represents the claims of experience. This conflicts sharply with the principle, basic to science, or to all rational cognition of the world, of the cognitive priority of reason over “experience”. Within cognition, it is reason that institutes or prescribes the forms of its own validation. Experiments, conceivedcognitively, represent the claims of reason. Thus, within the sphere of cognition or of the “theoretical idea”, as Hegel calls it, the whole notion of “experience” is an intensely problematic one. This is what the renewed disintegration of a more modern, historically-renovated empiricism expresses. In short, if there is an experimentalism of practice, this can no longer be secured by reference to the general contents of our cognition. It can no longer be secured by reference to “theory”, or to our “theories”, “hypotheses”, and so on.

Now turn to the second thesis.

“The question whether objective truth can be attributed to human thinking is not a question of theory”, not a question that can be posed purely by reference to the sphere of cognition, in the sense of “science”, or “theory”, “but is a practical question. Man must prove the truth, i.e. the reality and power … of his thinking in practice”.

That is to say, if human nature is rational nature, or if the “power” of thinking, i.e. of reason, imagination and so on, is really a species-power or an essential power, constitutive of human nature, then all this can only be established “in practice”, that is, by reference to human activity, or to the forms that this activity takes historically. And this is something Marx has already said. The content of the second thesis is already there,in nuce, in the statement, cited earlier, that “thepractical creation of the objective world…is proof that man is aconscious species-being”, that is, proof that humans have “consciousness”, or reason, as their intrinsic determination.

Experimentalism as the critical structure of practice

Deepened praxiologically, and thus shifted out of the closed circle of cognitive reason, the conception of “proof” or its implied experimentalism acquires a profoundly ontological and historical character. This is something that the narrow methodologism of Lenin’s “criterion of practice” fails to grasp. Its inevitable consequence is a collapsing of the second thesis into (a) a crude pragmatising “operationalism” which says that, unless our “ideas” or our “theories” can “work”, they are either meaningless (cf. Bridgman) or useless and therefore (cf. pragmatism of the classic variety) untrue, or (b) a philistine bourgeois utilitarianism of knowledge.206

The ideal of “experimentalism” which delegates to our experience of the world a critical function can only be grasped or understood properly beyond the sphere of rational cognition as such, where it is generally enshrined and from which it is then borrowed as a normative model or doctrinal assumption. In fact, matters stand quite differently. The experimentalism of cognitive reason, of science, is an experimentalism that glides continually into a tautology of reason. Or, within the sphere of science, “experience” is something absolutely inseparable from the rational structures that install it, or from the “reality” that reason institutes, or establishes.

But, if we turn now from the philosophy of science as such to the science of human intelligence in general, to “genetic epistemology”, the equivalence of experience and practice that Lenin and Della Volpe attempted to establish can be re-established in a more precise or rigorous manner, but from the opposite end, so to speak. Whereas Lenin and Della Volpe recast the function of practice as the traditional, empiricistic function of “experience” (i.e. as the transcendent and neutral arbiter of our conceptions of the world), Piaget’s genetic epistemology recasts the role of “experience” in the role that Marx ascribed to “practice”.207He establishes a principle first articulated, in an explicit way, by of Gramsci, which we can call the principle of the “practicity” of all human knowledge or intelligence. Gramsci made the mistake of arguing for this principle, a fundamental one, as he realised, by reference to, or on the normative model of, scientific experiments. He wrote,

“Scientific experience is the first cell of the new method of production, of the new form of active union between man and nature. The scientist-experimenter is also a worker, not a pure thinker, and his thinking is continually checked by practice and vice versa, until the perfect unity of theory and practice is forged”.208

It should be obvious that this entailed, in Gramsci’s case, as much as it did in Della Volpe’s, certain basic empiricist assumptions about the nature of science. What makes this less evident in Gramsci’s case is that this empiricism was presented as a radical and thorough-going historicism of knowledge. Thus, science, to Gramsci, is “the union of an objective fact with an hypothesis or a system of hypotheses that transcend the pure objective fact”. Again, “hasn’t all scientific progress shown itself, to date, in the fact that new experiences and observations have always corrected and developed earlier experiences and observations?”. All scientific knowledge is fallible and transitory, Gramsci argues, by virtue of the historical character of science itself, its location within the “superstructures” of society, or by virtue of its “practicity” which, to Gramsci, cannot be understood apart from its “historicity”.209

But, if we strip the argument of its mystical shell, its rational core consists in the notion of the “practicity” of all human knowledge. Now, as I said, this is a notion central to the modern study of intelligence initiated in that specific form by Piaget. The notion of the “practicity” of human intelligence implies, in Piaget’s work, two others that are inseparably connected to it. First, Piaget validates the “genetic” principle that learning is subordinate to development, i.e. that it can be understood in terms of an intrinsic psychological rationality which in Piaget, however, is conceived too narrowly, or purely in terms of his famous “sensori-motor” schemata. This “genetic” principle is one that could be directed as much against empiricist accounts of intelligence (e.g. the early associationist theories) as against the logical idealism of the Würzburg school. But, secondly, Piaget validates a much more subtle conception of the nature and role of “experience” in this process of development. If experience as such affects the process, then it does so mainly as the child’s own “experiments” with its environment, or as its “groping accommodations” to it. This is only the special case, or model-type, of the “practicity” of human intelligence, of the fact that the world that a child becomes “conscious” of is a world that the child has “constructed”, in practice and through continuous adjustments and modifications that manifest a distinctly experimental character.

Thus, Piaget will distinguish from the notion of “physical experience”, which is itself something elaborated practically, that is, the sort of experience that involves contact with objects and the gaining of knowledge by abstraction from the object itself, “logico-mathematical experience”, which likewise involves contact with objects, but

“By obtaining knowledge of these actions themselves and not from the objects as such”.210

The structures of human intelligence thus arise through a gradual “interiorisation of actions”, or they arise in a “practical” way. In the later phases of the “sensori-motor” period this is expressed by an “active experimentalism”. How, for example, does a child that is, say, less than a year old “learn” to draw an object through the vertical bars of its playpen? Piaget, analysing this case in an early work, concludes – through a gradual process of successive modifications, or by “the discovery of new means through experimentation”.211 Where gestalt psychology might have explained this process through the postulate that the child’s perceptual field is suddenly reorganised, Piaget answers, in a statement that is imbued with philosophical significance, “It is the action which fashions the field of perception and not the reverse”.212

Thus, Gramsci’s axiom must be reversed to be understood correctly. To him, the “unity of theory and practice”, or the “practicity” of knowledge, was the axiomatic epistemological foundation of Marxism. Of course, Gramsci drew inspiration for this idea from the Theses themselves. But, as I said earlier, he saw in the experimental activity of science itself the model of this “unitary process”, failing to see that life, or human activity in general, or general social practice, or historical self-activity, are the modelspar excellence of his experimentalism. In this sense, Lukács was much closer to the truth in seeing in the experimental activity of science the pure model of a “contemplative” stance to reality. For,

“the experimenter [in science – JB] creates an artificial, abstract milieu in order to be able to observe undisturbed the untrammelled workings of the laws under examination… He strives as far as possible to reduce the material substratum of his observation to the purely rational ‘product’, to the ‘intelligible matter’ of mathematics”.213

Thus, whereas Lukács is correct on a matter of detail, Gramsci is correct, and more profound, in the general conception. Whereas Lukács grasps the rational tautology of cognitive experimentalism, Gramsci understands the notion of experiment itself to comprise a deeper praxiological significance. The antagonism, or the tension, latent in Renaissance experimental rationalism is here externalised as a conflict between two distinct and opposed conceptions of the “experiment”. Lukács holds the rational tautologism of science against science, whereas Gramsci sees in its apparent or false tautoheterology the strongest point in its favour. That this tautoheterology is misconceived by Gramsci as a real dialectic of reason and experience is beside the point, or not of great importance, when measured against his intuitive grasp that the principle of experimentalism represents a fundamental feature of human activity as such.

But this is a notion that will remain merely “intuitive” in the Marxist movement. That it underlies the movement and shapes its conception of politics, of the self-activity of the class, is indicated sufficiently by the following passages, as a sample.


“The working class did not expect miracles from the Commune… They know that in order to work out their own emancipation … they will have to pass through long struggles, through a series of historic processes, transforming circumstances and men… In the full consciousness of their historic mission, and with the heroic resolve to act upon it, the working class can afford to smile … at the didactic patronage of well-wishing bourgeois-doctrinaires, pouring forth their ignorant platitudes and sectarian crotchets in the oracular tone of scientific infallibility”.214

Again, much more explicitly,

“The proletarian revolutions, like those of the 19th century, criticize themselves constantly, interrupt themselves continually in their own course, come back to the apparently accomplished in order to begin it afresh, deride with unmerciful thoroughness the inadequacies, weaknesses and paltriness of their first attempts.”215

Much later, in early articles from 1916, Gramsci will attack the tendency, common in Turin, to oppose “culture and intellectualism” to “practice and the historical reality by which the class is preparing the futurewith its own hands”. “Every revolution has been preceded by hard critical thinking”, he says, and by this “thinking” he means the sort of “introspection” that is embedded in the activity of the class at definite phases of its development.216


“The working class demands the right to make its mistakes and learn in the dialectic of history”.217

That is, learn actively, and thus experimentally, or in practice, or “practico-critically”.


“I vividly recall my “first experiment’ which I would never like to repeat…”.

Lenin again:

“In a revolutionary epoch like the present all theoretical errors … of the party are most ruthlessly criticised by experience itself, which enlightens and educates the working class with unprecedented rapidity”.


“Will Kautsky dare to mount a horse before he has learned to sit firmly in the saddle and to guide the animal in all its steps? We have reason to believe that Kautsky would not make up his mind to such a dangerous purely Bolshevik experiment… For the fundamental Bolshevik prejudice is precisely this: that one learns to ride on horse-back only when sitting on the horse”.218

Now precisely this is what Marx articulates in his conception of “production” in general. As the ontologically basic form or type of social practice, it is production that draws out most sharply the tautoheterology of reason. In labour, man “develops his slumbering powers”.219 Production in general, or production “for its own sake”

“means nothing but the development of human productive forces, in other words the development ofthe richness of human nature as an end in itself.”220

This nature, as a rational, creative nature, objectifies itself, in production, as “knowledge”, so that, in the more developed stages of social production, under capitalism, production itself signifies only the “power of knowledge, objectified”,221 or the production-process becomes a “scientific process,”222 or the activity of production becomes “practice, experimental science, materially creative and objectifying science”.223

The critique of capitalist society

Capitalism accomplishes this integration as disintegration. If the “practicity” of human intelligence, that “materially creative” force resident in human nature as such, is the typical symbol of human creativity, or of “reason”, then the “estrangement” of this capacity is the deepest symbol of human enslavement, or of man’s degradation or his “dehumanisation”.

The force of Marx’s critique of capitalism then rests precisely on the speculative or philosophical component in his conception of “production”. Acritique must be an argument “ad hominem”,224 it must make its demonstration in terms that are directly, and profoundly, relevant to human beings as the bearers of certain implicit powers.

In this sense, the analysis of the mechanism of capitalist production elaborated in Capital would come to form thescientific basis of acritique that had essentially already been summarized by Marx in the notion of “estranged labour”. The critical force of Marx’s demonstration lies in the general argument that while capitalism as a historical form of production, a specific social form of practice, creates unprecedented material and social conditions for the accelerated development of humankind, of human powers and capacities, it condemns humanity itself to “wage labour”. Thus the really “critical” concept in Marx’s demonstration lies not in the notion of “exploitation”, but in the more general conception of the bourgeois “division of labour”. Capital condemns humanity to a social division of labour whose starting-point is precisely the separation of “theory” and “practice”, as we can now call these moments. Theconstitution of this difference flows from the destruction of that integrally unified, “practico-critical” work-activity which before capitalism achieved its most complete, but historically limited and thus one-sided, expression in medieval craftsmanship.

Estranged labour, or wage labour in its deeper speculative conception, reflects a system of production, or a division of productive activity, in which

“the activity of the worker is not his own spontaneous activity”.225

That is, it is not an expression of the “creative”, rational powers that constitute the human essence. It is not an expression of the worker’spurposes, or her freedom. Thus, the worker can only experience her labour-activity as an alienation or an estrangement of her creative powers and her intentions. She can only experience labour as a silentcompulsion on her will. The social division of labour that uproots this “spontaneity”, or subordinates it in the form of compulsion, “separates theintellectual powers of production from manual labour”,226and converts those powers into the “might of capital over labour”. The introduction of machinery and the ever-increasing rationalization of the machine-based work-process, likewise a compulsion forced on each individual capital by the immanent laws of capital as such, doom an ever larger mass of human beings to endless routinism, or “drudgery”, that

“confiscates every atom of freedom, both in bodily and in intellectual activity”.227

The relationship of the worker to her activity becomes

“activity as passivity, power as impotence, procreation as emasculation”.228

But, if the rational spontaneity or material creativity of man is what constitutes him apart from the rest of nature, as subject, then this estrangement of the activity of labour threatens to abolish the distinction between humans and animals:

“The result is that man (the worker) feels that he is acting freely only in his animal functions … in his dwelling and adornment, while in his human functions he is nothing more  than an animal”.229

But, of course, this process begins earlier, or deeper, not in the factory, but in the home, not in the direct site of labour-activity but in the sphere of the “animal” reproduction of labour-power. The process begins with the working-class child, or with the worker as a child, or with those children who will over time form the working classes. The restricted, cramped, animalised development of the wage-labourer is a restricted, cramped, animalised development of his children.

If the basic characteristic of human development lies in the active or practical development of human powers, or of that “energy which is an end in itself”, or if the spontaneous experimentalism of human practice defines this practice as rational-creative or practico-critical activity, as the activity of a subject, then the “passivity” or the “impotence” of wage-labour activity forms the basic characteristic of estrangement.

Note how the city-form, central to the expansion of bourgeois civilisation as the unifying base and site par excellence of all its commodity-activities, evokes in the aesthetic representations of this civilisation the quality of being like a “dream”, or like “death”, or the image of something “sunken”, “hellish”, “unreal”.

“Activity as passivity, power as impotence…”

Marx writes, and someone else then echoes this as

“Shape without form, shade without colour,/Paralyzed force, gesture without motion”.230

Note how, as early as the 1840s, when the new industrial civilisation is still rough, amorphous, or only crystalline, the “crowd” acquires a sort of inner symbolism, from Engels through Poe into Baudelaire. The city is the base of this crowd, its specific element, the sphere in which it subsists. And the crowd is, for its part, the typical symbol of “active passivity”, of “gesture without motion”. In sum, the crowd expresses, in a peculiarly concentrated form, the automaticity that becomes the pervasive feature of the social life-process, or life practice, under capitalism.

If the worker’s activity is not “his own spontaneous activity”, it is not the activity of a subject, but automatic activity or activity established and imposed and coordinated as a function of machine-based industrial processes. It is interesting that when he analyses the “crowd” theme in Baudelaire, Benjamin should write,

“Independently of the worker’s volition, the article being worked on comes within his range of action and moves away from him just as arbitrarily. ‘Every kind of capitalist production … has this in common’, wrote Marx, ‘that it is not the workman who employs the instruments of labor, but the instruments of labor that employ the workmen. But it is only in the factory system that this inversion for the first time acquires technical and palpable reality’. In working with machines, workers learn to coordinate their own ‘movements to the uniform and unceasing motion of an automaton’.”

And Benjamin adds,

“These words shed a peculiar light on the absurd kind of uniformity with which Poe wants to saddle the crowd – uniformities of attire and behavior, but also a uniformity of facial expression”.231

When, in the manifesto that Marinetti published around the turn of the century, announcing Futurism, he declared, “we will extol immense crowds, moved by work, pleasure or rebellion; the multicolored and polyphonic fits of revolutions in all modern capitals”,232 then, in this futurist image, he simply drew the contrasting picture of the present, that is, of a bourgeois civilisation now entering its phase of cultural decadence, of crowds not moved by “work”, of “capitals” yet to witness “revolutions”.

“‘All machine work requires early drilling of the workers’. This drill must be differentiated from practice. Practice, which was the sole determinant in craftsmanship, still had a function in manufacturing. With it as the basis, ‘each particular area of production finds its appropriate technical form in experience and slowly perfects it’… On the other hand, this same manufacturing produces ‘in every handicraft that it seizes upon a class of so-called unskilled labourers. If it develops a one-sided speciality into a perfection … it also begins to make a speciality of the absence of all development’”.233

“The drill must be differentiated from practice”, Benjamin writes. So must habit, routine, discipline, repetitive, monotonous, one-sided specialisation. Recall Engels on the Renaissance,

“the heroes of that time wore not yet in thrall … that is crushed by, enslaved to, the division of labour”.

The civilisation of the Renaissance, based on a more profoundly unified, integral connection between “intellectual” and “manual” labour, a connection symbolised by its technical and scientific interests and by its specific form of manufacturing, was also the civilisation that could articulate well in advance of Marx the rationalist humanism that Colletti perceptively brings back into focus. See how, in Cusanus, all the dialectical and rationalist elements of Marx’s humanism are already prefigured in the notion that,

“The active creation of humanity has no other end than humanity itself. For humanity does not proceed outside itself while it is creating, nor does it produce anything new. Rather, does it know that everything it creates by unfolding was already within it”.234

In this notion, Cusanus will “praxiologise” Plato’s conception of “learning” as a process of “recollecting” what already lies “implicit” within us.

It follows that, when Marx writes about the activity of production divested of the specific social forms that it takes under capitalism, or of the production-activity of a communist society, his emphasis centres on the reintegration of the “intellectual” (creative, rational, experimental) aspects of production with the process of labour as such. It follows that he will see in this production the “development of that energy (that power or fund of power) which is an end in itself”, and write that this can only be the case when labour “is of a scientific character, as well as being general labour”.235 Scientific, that is, rationally creative, experimental, critical, in other words, subjective, not

“the exertion of a specifically trained force of nature but the labour of a subject”.

In short, Marx’s theory of “estranged labour” represents a praxiological critique of capitalism.

“Theory” and “practice”

The reverse side of the process by which the social division of labour dehumanises production is the forcible separation and seclusion, into a functionally specialised and self-contained sphere, of the critical, rational, experimental, scientific, and technical moments of social practice. These moments are thus “preserved” and functionally consolidated, but only abstractly, or one-sidedly, and thus in forms that are necessarily specialist, esoteric, abstract, and utopian.

All “theory” bears the imprint of this separation. Or, it is this separation, a real historic process, that brings theory into being. The separation, or divorce, or hiatus, or disunity of “theory” and “practice” is thus something intrinsic to capitalist society, to the division of labour that it generates and reproduces, and a truly permanent or lasting or final integration of those moments becomesimpossible within the historical limits of capitalism.

“Abstract thought” and “sensuous contemplation” thus form the twin antipodes of the social life-process under capitalism. Sensuous contemplation, for

“As labour is progressively rationalized and mechanized, the worker’s lack of will is reinforced by the way in which his activity becomes less and less active and more and more contemplative.”236

Or, as Engels writes precisely in the year that Marx composed both the Manuscripts and theTheses,

“As voluntary, productive activity is the highest enjoyment known to us, so is compulsory toil the most cruel, degrading punishment… The division of labour has multiplied the brutalising influences of forced work. In most branches the worker’s activity is reduced to some … purely mechanical manipulation, repeated minute after minute, unchanged year after year… What abilities can a man retain in his thirtieth year who has made needle points or filed toothed wheels twelve hours every day from his early childhood?… The worker’s activity … offers no field for mental activity”.

“The supervision of machinery, the joining of broken threads, is no activity which claims the operative’s thinking powers… There is no better means of inducing stupefaction than a period of factory-work, and if the operatives have, nevertheless, not only rescued their intelligence, but cultivated it and sharpened it … they have found this possible only in rebellion against their fate and against the bourgeoisie… Or if this indignation against the bourgeoisie does not become the supreme passion of the working-man, the inevitable consequence is drunkenness”.237

Against which stands the “abstract activity of thought”, expression of the same underlying social movement:

“Division of labour only becomes truly such from the moment when a division of material and mental labour appears. From this moment consciousness can really flatter itself that it is something other than consciousness of existing practice…From now on consciousness is in a position to proceed to … the formation of ‘pure’ theory”.238

Thus “theory” is constituted by the social division of labour, that is, the disintegration of work as a unified activity. The monotonous, stupefied, contemplative character of “manual”, purely sensuous, physical labour in the factories is reflected, in reverse, in the abstract, specialist, utopian character of “theory”. “In Germany”, Marx writes, “practical life is as devoid of intellect as intellectual life is of practical activity”.239

Now, Marxist theory, the “theory” of the working-class movement, no more escapes these determinations than any other sphere of abstract, specialist “intellectual life”. In relation to the social life-process and practice of bourgeois society as a whole, this theory necessarily retains the twofold characteristics of something “abstract”, specialist, esoteric, and of something “utopian”, other-worldly. As a “theoretical idea” or as cognition or science, it is a system of thought permeated by the “finitude” of having an otherness, an objective world over against it, closed in on itself. The striving for “concreteness” that characterises the Notion as “subjective” Idea, as which it is only “theoretical”, is realised by it in such a way that “the Idea in the first instance only gives itself a content whose basis is given”. Or, in “the theoretical idea”, in all theory, “the subjective Notion … stands opposed to the objective world from which it takes into itself a determinate content and filling”. “Accordingly, cognition still retains its finitude in its realized end; in its realized end it has at the same time not attained its end, and in its truth has not yet arrived at truth”.240

The Marxist movement was, to whatever degree, and however unselfconsciously, conscious of this, of its “finitude”. In Kautsky, and the positivising currents of Marxism generally, it would find the positive assertion of this finitude in the thesis that “socialism and the class struggle arise side by side and not one out of the other; each arises under different conditions”, that is, under the silent pull of the social division of labour, so that “Socialism”, the theory of the movement, arises in circumstances no different from those that determine the other specialised spheres of bourgeois intellectual activity.

“Modern socialist consciousness can arise only on the basis of profound scientific knowledge. Indeed, modern economic science is as much a condition for socialist production as, say, modern technology… The vehicle of science is not the proletariat but the bourgeois intelligentsia”.241

A positive exposition, for Kautsky ignores or fails to draw out the historically limited and alienated character of this separation. It follows, conversely, that the “class struggle” is, on the other side, accepted or endorsed implicitly as “practical life devoid of intellect”. It follows, furthermore, that the independent activity or the self-activity of the class, which, as Engels says, can only assert itself or express itself as anopposition to the rule of capital, given the nature of capitalist production and its stupefying effects, must then bedowngraded to a merely “blind” spontaneity, to something merely, or purely, “instinctual” or “unconscious”. And this, then, is a line of thought that Lenin develops in 1903 on the basis of Kautsky himself.242

        It is a line of thought that even Gramsci, despite the profoundly praxiological character of his thought,243 and despite his deeper historical insights, does not completely escape. For Gramsci will likewise start from the historically given separation of “theory” and “practice” and understand the process of their “unification” as a fusion of the intellectual strata with the popular masses, or as a process whose “true site is the Party”.244 It is a line of thought that Luxemburg alone refuses to accept, and this is what constitutes the real specificity of her Marxism, but one that she will fail to overcome decisively. For, where Kautsky and Lenin positivise the separation of “theory” and “practice” in the idea that “socialist consciousness” is brought to the proletariat “from the outside”, that left to its own “spontaneity” the class will fall victim to “bourgeois ideology”, much as Christianity thought that a peasantry without the Bible was a peasantry steeped in drunkenness, Luxemburg, for her part, simplyaccommodates to this separation by ascribing to “pure theory” and the “class war” different tempos, even different concerns, so that “theory” always “offers” to the class struggle much more than the struggle is actually capable of absorbing, and so that all problems that “are important from the outlook of pure theory” are “comparatively unimportant from the practical outlook of the class war”.245

Thus, starting from the absolute and real, historically given, separatedness and fixity of “theory” and “practice”, even the most enlightened sections of the Marxist movement will conceive of their “unification” as an external process, or avoid the problem altogether (as Luxemburg did). In short, almost a century after Hegelianism first confronted and attempted to resolve this, the Marxists will have advanced their conception of the problem no further than the point it reached in Moses Hess. For, already in Hess,

“the duality of theory and practice assumes the form of a duality between the historical movement whose ‘mission’ is to bring socialism about … and the philosophical theory of this movement which is supposed to give it clarity and direction and explain its real goals to it”.246

Against all of which, Marx writes that

“Theory is realized in a people only insofar as it is a realization of people’s needs”.247

If one objects that this surely is how Kautsky, Lenin et al. also see it, that in their conception, too, revolutionary theory is directed at the “realization of people’s needs”, one misunderstands Marx. In the positivising conception of a theory that is “instilled into” the class, the class, its movement, its practice, its self-activity, even its spontaneity, become realisations of that theory. And this is the reverse of Marx’s meaning. Marx is saying,theory must be a realisation of “people’s needs”. That is, “it isnot enough that thought should striveto realize itself”, or, it is not enough that “theory” should strive to find in the activity and movement of the class the principle of its own realisation, or that revolutionaries must seek to embody their programmes, their ideas, their conceptions in the activity and struggles of the class, but “reality must itself strive towards thought”.

And what does that mean if not that this “reality”, the practice of the class, its self-activity, its movement, its struggles, must strive to recover their immanent moment of self-consciousness, of self-criticism, ofexperimentalism, or recover that moment which in the social division of labour generalised by capitalism is subordinated, crushed, driven underground, rendered merely implicit, taken out of production, externalised, specialised, set apart? What does Marx mean except that the class itself, humanity at large, must come to a realisation, through itself, of its own “powers” or “capacities”, which arehuman powers and capacities, and which as specificallyhuman powers and capacities, are rational powers, powers of creativity, of thought, of imagination, of freedom, of the spontaneous exertion of will in accordance with rational purposes? In short, that the mass of workers must “learn”, recollect that which already lies implicit within them as intrinsic determinations of their human nature, and that they can only “learn” through their own practice, or in their own struggles? For,

“men who do not feel themselves to be men accumulate for their masters like a breed of slaves or a stud of horses”.248

A “reality that strives towards thought” is a reality, a practice, a movement of life in which the mass of humanity recovers its “self-esteem” or its “sense of freedom”, rediscovers itself as the bearer of certain powers or capacities that the social forms of life-practice fetter, constrict and uproot. And this rediscovery, or recollection, or learning, or this self-consciousness is something that cannot arise “abstractly”, or from the “outside”, but only through workers themselves, through their own activity and struggles. And that is precisely why “every step of real movement is more important than a dozen programmes”.249

For it is a movement of reality itself towards thought, it is practice recovering its implicit moments of “criticism”, or it is the class educating itself in the only way that is possible for it, that is, actively.

Thus, if there is a “true site” of the unification of “theory” and “practice”, then this cannot be the party, as Gramsci claims, it can only be the sphere of class-activity, class-practice, class-education, or self-education and struggle.

And it follows also that a revolutionary “theory” that cannot assist this process, that cannot form an experimental reflection of the activity and movement of the class is notrevolutionary theory. As this experimentalism of class activity, or as this “criticism” of class activity, revolutionary theory can only be a “demonstrationad hominem”, that is, a critique of the existing social forms of practice in terms of the powers, capacities, dispositions that define humans, and an attempt to stimulate the activity of the class in preciselythose directions that allow it to discover its own capacities, which arehuman capacities, and in this way, slowly, through a protracted process, to rediscover their “self-esteem” and their “sense of freedom”.

It has hardly ever been noticed that in the Critique of the Gotha Programme Marx, to give one example of this trend of thought, of this conception of politics, opposed the demand for the abolition of child labour as “reactionary”. On the grounds precisely that “an early combination of productive labour with education is one of the most potent means for the transformation of present-day society”.250

Now this general and abstract definition of the role of revolutionary theory implies quite exactly that a theory that does not conceive of its own development as a practical process, as an expression,experimental,critical, rational and so on, of thereal life-process and activity of the class, or a theory that does not conceive of its own development as a function of that “reality striving towards thought”, remains a mere theory, something merely “abstract”, specialist, utopian, something “contemplative” in its own way, permeated and infected by the “finitude” of all cognition or of all purely cognitive reason.

This implies, more concretely, that, while at the level of its most general conceptions or general premises, a revolutionary theory, to be theory at all, must fulfil the positive cognitive function of reason, that is, must validate in its own way, or for its own object, the cognitive priority of reason over perception, or must proceed from certain firm,scientific principles, the actual process of development of such a theory, that is, the process that makes it no longer only a theory, but arevolutionary theory that is no longer merely abstract, contemplative, specialist, but revolutionising, this process can only be a practical-experimental one, and this through the inner connection of these moments. It can only be “practical” as a function of itsexperimental mediations of class activity and class struggle, and it can only be this experimental mediation of class activity if it conceives its own development experimentally, practico-critically, or as a validation, in activity or struggle or history, of those moments that lie only implicit within it as its inner, general, abstract, purely rational determinations. Thus a revolutionary theory can only develop, that is, become revolutionary, by drawing out its inner potential, and this it can only do practically and experimentally, in a protracted process which is a process ofknowledge, but no longer “abstract” knowledge, but knowledge that arisespractically as the “self-mediating unity of consciousness and reality”.

Conceived as such a process, i.e. as a development, “theory” breaks down into an infinity of conjectural, experimental, critical moments whose inner tendency is established firmly, but only abstractly, or tautologically, by its own general, rational, scientific principles. Only this conception, tautoheterological in its essence, can account for, validate, both the dogmatism that characterises all science, if it is truly scientific and not a show of science (cf. hypotheticism), which is the attitude by which all cognitive reason holds by its general conceptions despite the contradictory evidence of the senses, or perception, or impressions, etc.,and account for and validate the dynamism, the flexibility, or the openness that is likewise the hallmark of critical thought. Or only this conception, tautoheterological, can avoid theabstract dogmatism of a thought closed in on itself, the latent tendency of philosophical rationalism, and the shalloweclecticism, or impressionism, or empiricism of a thought that abandons its general premises on the slightest encounter with the contradictory “data” of “experience”.

But, again, at this level too, the Marxist tradition, just as it proceeded from the fixity and separatedness of “theory” and “practice”, here also polarised its own development into the distinct and opposed moments, thoroughly one-sided, of abstract dogmatism, or tautology, and shallow eclecticism, or induction. If we look back over the history of the Marxist movement, then precisely this is the most general motor of its theoretical development. Doctrinalism, or the abstract tautologising dogmatism of the general, scientific principles themselves, calls forth as its one-sided antithesis, inductivism, or the appeal to “experience”, to “facts” that are contemplated without the mediation of any principles. Such is the nature of the first struggle in German Social Democracy, the one that opposes “orthodox” Marxism to Bernstein. Such also is the nature of the early splits that break the unity of the Fourth International and account for its subsequent disintegration. No one can deny that, in both periods, the general state of the working-class movement and the deeper movements of the social life-process as a whole must be taken into account to understand those struggles historically. It is no accident that the 1890s and 1940s saw capitalism emerging, in both periods, renovated and restructured after protracted periods of major depression. But we cannot proceed from these general historical determinations to the nature of the struggles and crises within the Marxist movement without mediating the explanation through the very nature of “revolutionary theory” in these periods. It is a striking fact, for example, that these specific periods were either immediately preceded or followed by a tremendous upsurge in the self-activity of the class. Yet, in the first, Luxemburg remains literally the only Marxist to appreciate the significance of this upsurge, or to say anything at all significant about it. In this sense, the crisis of German Social Democracy, whose final manifestation comes in 1914, is repeated, in a different way, in the history of the postwar left. For if the strikes of the 1890s attracted Luxemburg’s attention, and the upheavals following the First War found their experimental resonance in the early writings of Gramsci, in Turin, then the postwar left has yet to show asingle example of any effort to comprehend the rich history of the movement of the class from the 1930s down to today. A theory that cannot become the critical or experimental reflection of the class is a theory that remains abstract and utopian. And the price that revolutionaries pay for this “abstractness” when they enter the sphere of struggle, when they seek to integrate with the activity of the class, is pragmatism. The abstractness of theory, its tautological dogmatism, and the pragmatism of politics thus become expressions of the same movement, reverse signs of the same conception of politics, now no longer merely proceeding from the separation of “theory” and “practice” but shaping the content of its conceptions within the framework of that separation.


J.B., March 1977

Image “Protestas en Chile” bytodosnuestrosmuertos is licensed underCC BY-NC-SA 2.0


  • 1.F. Engels, “Ludwig Feuerbach and the End of Classical German Philosophy”, from the foreword to the 1888 edition, K. Marx and F. Engels, Selected Works, vol. 3 (Moscow, 1973), p. 336.
  • 2.V. I. Lenin, “What the ‘Friends of the People’ Are and How they Fight the Social-Democrats”, Collected Works, vol. 1 (Moscow, 1963), pp. 136 ff.
  • 3.Max Adler, La pensée de Marx, cited Lucien Goldmann, “Propos dialectiques: y a-t-il une sociologie Marxiste?”, inRecherches dialectiques (Paris 1959), p. 289. Adler’s conception is of course blatantly in conflict with Marx’s own repeatedly affirmed views about Comte, cf. the letter to Engels, 7 July, 1866, where he says that “compared with Hegel” Comte is “wretched”, and refers to his “trashy positivism”, or the letter to Engels dated 20 March 1869, where he says, “Positive Philosophy means ignorance of everything positive”.
  • 4.N. Bukharin, Historical Materialism: A System of Sociology (London, 1925),passim.
  • 5.K. Marx, “The Difference between the Democritean and Epicurean Philosophy of Nature” (Marx’s doctoral dissertation), Collected Works, Vol. 1 (Moscow 1975) p. 40, “Democritus, for whom the principle does not enter into the appearance, remains without reality and existence, is faced on the other hand with theworld of sensation as the real world, full of content … it is the unique real object and as such has value and significance. Democritus is therefore driven intoempirical observation. Dissatisfied with philosophy, he throws himself into the arms ofpositive knowledge”.
  • 6.This tension defines the whole of Colletti’s Marxism and accounts thereby for its freshness and fertility. The citation is from “Marxism and the Dialectic”, New Left Review, 93, p. 29, where he adds, “I do not know whether the existence of these two aspects is fatal or advantageous”.
  • 7. “Value Essence and Value Existence in Marx”, August 1976.
  • 8.Of course, this hostility is not supported equally well by the three of them. For example, Colletti derives, from Della Volpe, something like “philosophical erudition”, which neither Althusser nor Timpanaro show or want to acquire. The twenty-odd references to Hegel in On Materialism (London, 1976) contain not a single reference to any of Hegel’s own writings. This follows from the more consistent and thorough-going character of Timpanaro’s materialism, and from its reduction of the “problem of Knowledge” to physiology. Though Colletti (Marxism and Hegel, London 1973) and Timpanaro share basically identical polemical posturesvis-à-vis natural-scientific materialism, the only brand they represent, and what they see as the “intrinsically idealist” character of the dialectic, Timpanaro is more consistent in the view, not shared by Colletti, that epistemology is inherently “voluntarist”, or inseparable from an “idealist” outlook. Thus he writes, about the “problem of knowledge”,  “This implies a polemical position towards the major part of modern philosophy, which has entangled and exhausted itself in the setting up of ‘epistemological traps’ to catch and tame the external datum, in order to make it something which exists solely as a function of the activity of the Subject. It is important to realize that epistemology has undergone such an enormous development in modern thought because it has not only corresponded to the need to understand how knowledge arises, but has been charged with the task of founding the absolute liberty of Man” (On Materialism, p. 35). Naturally, with this sort of position, Timpanaro is diplomatically silent about theTheses on Feuerbach, and especially the first thesis. On the other hand, Colletti inherits from Della Volpe a far less polemical, more tolerant conception of classical philosophy, and a deeper acquaintance with it.

                Timparano is thus able to grasp the inconsistency in Colletti’s materialism. “What one cannot understand is why the affirmation of existence of a reality not reducible to thought … should be regarded in and of itself as specific proof of Kantianism” (op. cit. p. 78). Timparano is perfectly right. If the sole purpose of invoking Kant is to bolster an empiricist critique of Hegel with “classical” authority, or to reduce Kant to an empiricist (and Della Volpe was already conscious that this could not be done, hence his far more critical positions towards Kant), then perhaps Locke’s empiricism, or 18th century materialism might have done just as well (after all, it was to Locke that Marx and Engels turned by way of finding within the tradition of philosophy an authoritative critique of “metaphysics”).On the other hand, Timparano can only secure his revamped vulgar materialism through deliberate one-sided polemical exaggeration. He writes, “Perhaps the sole characteristic common to virtually all contemporary varieties of Marxism is their concern to defend themselves against the accusation of Materialism” (p. 29). But this is nonsense. For at least twenty years now, or, say, since Della Volpe’s first edition of Logica, the whole tendency has been towards materialism, towards its revival and greater sophistication, and against anything at all to do with the German philosophical tradition. In this sense, despite his philosophical apathy or even philistinism, Anderson (Considerations on Western Marxism, London 1976) is more correct when he says that hostility to “Hegelianism” has been the unifying bedrock of this turn to materialism: “Althusser’s categories explicitly included Colletti in the Hegelian tradition he repudiated, while Colletti’s logic assigned Althusser to the Hegelian heritage he denounced” (Op. cit., p. 70).

                The real picture of this “contemporary” period, then, is that while some earlier pseudo-Hegelian forms of Marxist theory came back into currency (e.g. the revival of History and Class Consciousness, the renovation of Korsch, the renewed popularity of ‘Critical Sociology’), at the very same time there started a fairly systematic onslaught on Hegel himself. The two are by no means symmetrical phenomena: there was no “back to Hegel” movement over this period, but, rather, the rediscovery of certain Hegelian themes mediated through Lukács and Marx himself. On the other hand, the attack, launched initially in Italy and followed up in France, concentrated quite relentlessly on Hegel.

             This is what accounts for the lack of any real polarisation within the ambit of Western Marxism. Take the phenomenon that Anderson calls, in a purely descriptive way, “criss-crossing” (Considerations, p. 72). Both Colletti and Timparano represent identical movements of reaction to the “neo-Hegelian” permeations of Italian idealism and its legacy. Both defend a realist ontology, common sense, and natural science. Both repudiate the dialectic. But then compare their radically contrasting attitudes to Engels and Engelsism, the point over which Timparano polemicises against Colletti.Again, contrast the positions of Althusser and Timpanaro on their respective conceptions of “Method”, but note their fundamentally common attitude to Hegel and their paradoxical points of convergence, e.g. the dismissal of classical philosophy by both as pure illusion. Or, again, compare the retention of the “dialectic” in Althusser, the massive assault on “dialectical materialism” in Colletti, the revival of this dialectical materialism by Geymonat, and their common espousal of materialism. Or take this final example: the wholesale rejection of the theme of “alienation” by Althusser, its emphatic endorsement by Colletti, its peculiar evaporation in Timpanaro (where is the theme even mentioned?) and their common, unifying hostility to “Hegelianism”.Philosophically, then, the position of contemporary Marxism could not present a more striking contrast to the situation that prevailed around the 1890s.

  • 9.K. Marx and F. Engels, The Holy Family, or Critique of Critical Criticism (Moscow, 1975) pp. 147-48.  I have started with this text because philosophically it is one of the least evasive or most explicit in the entire corpus of writings of Marx and Engels. Many of the views expressed in it were later rescinded or revised. The notable case of this is the “great admiration” Marx would later affirm for Leibniz.
  • 10.J. Locke, An Essay Concerning Human Understanding, p. 59. Already by 1859, Marx had changed his view of Lockean empiricism, writing inA Contribution to the Critique of Political Economy (Moscow, 1970) that “Locke even demonstrated in a separate work that the bourgeois way of thinking is the normal human way of thinking” (p. 77), a profoundly sarcastic comment on empiricism.
  • 11.G. W. Leibniz, New Essays on Human Understanding (1700), inLeibniz: Philosophical Writings, trans. M. Morris (London, 1956), pp. 143f. Thus Leibniz says here that the senses are necessary for all actual knowledge of the world, or that the real process of cognition cannot dispense with perception. For the same point cf. I. Kant,Critique of Pure Reason (2nd edition, trans. Meiklejohn, London 1934) p. 86: “With respect to these conceptions, as with respect to all our cognition, we certainly may discover in experience if not the principle of their possibility yet the occasioning causes of their production (Gelegenheitsursachen ihrer Erzeugung). It will be found that the impressions of sense give the first occasion for bringing into action the whole faculty of cognition.” Thus the role of sense experience is here confined to the function ofAnlass, “occasion”, with which compare “stimulus” in the modern version of the “innateness” hypothesis argued by Chomsky, e.g., “Recent Contributions to the Theory of Innate Ideas”, inBoston Studies in the Philosophy of Science, vol.3, edited by R. S. Cohen and M. Wartofsky (Dordrecht 1967).
  • 12.M. Blake, C. J. Ducasse and E. H. Madden, Theories of Scientific Method: The Renaissance through the Nineteenth Century (Washington, 1960), p. 19. This book has a strongly empiricist bias, which is useful because it produces, unconsciously, a more balanced account of the classical rationalists, though at places (e.g. on Bacon, Hobbes) the discussion remains totally superficial.
  • 13.Blake et al., Op. cit. p. 8 for Vives, and p. 7 for Pico della Mirandola in the Platonist tradition. Pico based his attack on Aristotelian epistemology much more on the ancient-sceptical thesis of the intrinsic fallibility of our sense-perceptions.
  • 14.Bacon, Novum Organum, Book One, which deals with “method”.
  • 15.D. Hume, Treatise of Human Nature, Book 1, Part 3, Sect. 6.
  • 16.Kant, Prolegoma to any Future Metaphysics , ed. L. W. Beck (New York 1950) pp. 5ff., showing how deeply Kant himself was agitated by the problem of induction. He says, “Since Locke and Leibniz … nothing has ever happened which could have been more decisive to the fate of metaphysics than the attack made by it by David Hume”.
  • 17.Marx and Engels, Holy Family, pp. 147, 109; Marx, “Economic and Philosophical Manuscripts (1844)”, inEarly Writings , tr. R. Livingstone and G. Benton (Pelican Books, 1975), p. 281; cf. “Feuerbach is the only person who has a serious and critical attitude to the Hegelian dialectic and who has made real discoveries in the field” (p. 381).
  • 18.L. Feuerbach, Vorläufige Thesen zur Reform der Philosophie, cited A. Schmidt, The Concept of Nature in Marx (London, 1973), p. 24.
  • 19.Feuerbach, Op. cit. and inPrinciples of the Philosophy of the Future, cited Lukács, “Moses Hess and the Problems of Idealist Dialectics”, in Lukács, Political Writings 1919–29 (London, 1972) p. 204 ff. And cf. Lukács’ own critical comments at pp. 206ff.
  • 20.The real break here comes with Mach, not Comte, though Machism itself is prepared by the strong undercurrent of empiricism that swells up in the early and middle parts of 19th century. It is perhaps worth pointing out here that Kant’s First Critique played a significant role in this historical renovation of empiricism, specifically through Whewell’s writings on inductivism and in the distinction that the Scottish engineer William John MacQuorn Rankine would draw in the 1850s between the “hypothetical” and the “abstractive” methods in science. Here “hypothetical” is a resonance of Kant’s notion of the purely regulative function and character of the “ideas of reason” (cf.Critique, op. cit., “Appendix to the Transcendental Dialectic” pp. 373ff. and its basic proposition that “If we review our cognitions in their entire extent we shall find that the peculiar business of reason is to arrange them into a system”, and the whole discussion of this aspect of theCritique in H.W. Cassirer,Kant’s First Critique, London 1968, Chapter 14.) Rankine’s “abstractive” method, on the other hand, was a throwback to Kant’s phenomenalism, and in this form was picked up again by Kirchhoff and Mach, to become central in the conventionalism of Duhem and Poincaré. (For Kirchhoff, cf. Ludwig Boltzmann,Theoretical Physics and Philosophical Problems: Selected Writings, ed. B. McGuinness, Dordrecht 1974, p. 16: “In his comprehensive work on mechanics Kirchhoff very clearly sets himself as a task merely to describe natural phenomena as simply and perspicuously as possible, renouncing all explanation, and since then what in physics used to be called explanation has repeatedly been called a mere description of the facts”). The leader of Marburg neo-Kantianism, Hermann Cohen, will likewise transform the First Critique into the epistemological basis of positivism in science, cf.Kants Theorie der Erfahrung (Berlin, 1885).
  • 21.Feuerbach, Lectures on the Essence of Religion, cited in Lenin,Philosophical Notebooks, inCollected Works, vol. 38 (Moscow, 1972), pp. 81, 75.
  • 22.Feuerbach cited Schmidt, Op. cit. p. 25. Of course, this was the element in Feuerbach that exerted the deepest influence on Marx.
  • 23.Cf. Ludwig Büchner, Force and Matter (New York, 1891) p. 205, where Feuerbach is called “the philosopherpar excellence of emancipated and self-contained humanity”. But, to materialists like Büchner it was not Feuerbach’s humanism but his atheism that exerted the greatest appeal. Or, more precisely, his humanism was received as atheism. This accounts for the otherwise anomalous fact that. after praising Feuerbach for his “humanism”, Büchner goes on to collapse “human” beings into nature, in chapters that are simply unparalleled for their crudeness; for example, see his reflections on “madness” which begin, at p. 227, “Mad people, and people who are mentally diseased, always have diseased brains”.
  • 24.Cf. Engels, “Ludwig Feuerbach”, op. cit., p. 349.
  • 25.Ibid. p. 344. Note carefully, “We all became Feuerbachians”.
  • 26.Feuerbach himself divided philosophy, historically, into “idealism” and “realism or empiricism”, identifying with the latter. The “later tradition” refers specifically to Lenin but also to Colletti.
  • 27.Cited Schmidt, op. cit., p. 26. The whole notion of “sensuous objects”, so central to the argument in theManuscripts, is Feuerbachian through and through, see Colletti,op. cit. p. 11 ff.
  • 28.Feuerbach cited in Lenin, Materialism and Empirio-criticism (Moscow, 1967), p. 188.
  • 29.Engels, Op. cit., p. 348; Lenin,op. cit., p. 188.
  • 30.Early Writings, p. 355; a view that Marx will later emphatically reject.
  • 31.Holy Family, p. 154.
  • 32.Incorrect, because Bacon says something quite different, e.g. “To the immediate and proper perception of the Sense I do not give much weight, but I contrive that the office of the Sense shall be only to judge of the experiment” (Novum Organum, 1. 98, and his “Plan” for The Great Instauration.)
  • 33.Holy Family, p. 150f. All this may be due to Engels, but it is pointless arguing like that. Marx himself much earlier (e.g. 1842) had a far more sophisticated conception of the whole matter. Cf. as one example of this, “The Supplement to Nos. 335 and 336 of the Augsburg Allgemeine Zeitung on the Commissions of the Estates in Prussia”,Collected Works, op. cit., p. 295: “one would have to demand of the author that he should make a more thorough study of Nature and rise from the first sensuous perception of the various elements to a rational perception of the organic life of Nature… Just as Nature does not confine itself to the elements already present, but even at the lowest stage of its life proves that this diversity is a mere sensuous phenomenon that has no spiritual truth, so also the State … must not and cannot find its true essence in a fact apparent to the senses”. It is tothis sort of conception that Marx will laterreturn in the 1850s.
  • 34.Holy Family, p. 152. But for Locke, see note 10 above. Their judgement about Hobbes is as strange as the conception of Bacon’s empiricism is wrong. For Hobbes inclined to an extreme form of rationalism, fairly unique in the English tradition, distinguishing “science” from mere “cognition”; cf. the famous passage in De homine,Opera Philosophica, vol. 2 (London 1839), p. 92: “Scientia intelligitur de theorematum, id est, de propositionum generalium veritate…Quando vero de veritate facti agitur, non proprie scientia sed simpliciter cognitio dicitur”, i.e. science (in thestrict sense) deals with propositions that can be demonstrated with mathematical necessity, and not with mere truths of fact. Further, note how Hobbes denigrated the role of experimental observations in the strongest possible terms: “If experimentations of natural phenomena are to be called philosophy [science – JB], then pharmacists are the greatest physicians of all”. Given all this, it is strange to say that Hobbes failed to “prove” Bacon’s empiricism!
  • 35.This is probably the only passage in Marx that directly attacks the notion of “innateness”. It is a view not at all easy to reconcile with his notion of “species-powers”, “drives”, “dispositions” and so on. Pure 18th century materialism.
  • 36.For Mach, cf. R. S. Cohen, “E. Mach: Physics, Perception and the Philosophy of Science” in Boston Studies, Vol.6: E. Mach Physicist and Philosopher (Dordrecht, 1970), p. 130: “To Mach metaphysics is an illusion of knowledge, a deception of the human spirit”. After Feuerbach and Mach, European thought will have to wait for the young Wittgenstein before any equally destructive idea of “metaphysics” is again proposed.
  • 37.Dewey apud Della Volpe,op. cit. infra p. 134. Here the basic error of “idealism” is located in “the hypostatization of a logical function into a supra-empirical entity” (Dewey’s terms).
  • 38.P. Frank, Modern Science and its Philosophy (Harvard 1950), with useful biographical details on the formation of the Vienna Circle.
  • 39.Frank, Op. cit. pp.174–5, referring to Cassirer’s later work,Determinismus und Indeterminismus in der modernen Physik (Göteborg, 1937). Colletti,Marxism and Hegel, Ch. 7 will likewise use Cassirer against “metaphysics”. On the other hand, it only shows Cassirer’s deep philosophical ambivalence that Della Volpe can criticise him precisely for his “excessive” rationalism.
  • 40.The programme was called “Physicalism”. On its disintegration cf. M. W. Wartofsky, “Metaphysics as Heuristic for Science”, Boston Studies in the Philosophy of Science, vol. 3 (Dordrecht, 1967), p. 123f. Craig’s effort was directed towards building a system containing only “observational terms”. This, he found, was possible but at the cost of an infinite set of postulates. For Craig’s theorem called the “eliminability theorem”, see “On Axiomatizability within a System”,Journal of Symbolic Logic, 18 (1953) and his later paper “Replacement of Auxiliary Expressions”,Philosophical Review, 65 (1956).
  • 41.On this aspect of Wittgenstein’s brief flirtation with logical positivism cf. P. Hacker, Insight and Illusion (Oxford, 1975), p. 104 ff., “experiential propositions which are not directly and conclusively verifiable by reference to phenomenal experience are hypotheses… Propositions about the past, about laws of nature, etc., are hypotheses”.
  • 42.Cited Lenin, Philosophical Notebooks, p. 68. “Convenience” was, of course, a central notion in Mach’s theory of explanations in science, and likewise basic to Duhem’s conventionalism. Thus inThe Aim and Structure of Physical Theory (New York, 1962), Duhem saw the sole function of “abstract” reasoning in science as one of ordering or “cataloguing” the innumerable “facts” of our experience at successively higher levels of synthesis (laws, theories). In the medieval period conventionalism bore the less euphemistic name “fictionalism”. It was through this conception of the nature of our cognitive claims that Osiander sought to defuse the revolutionary implications of Copernican astronomy and to render them palatable to the reactionary, theological, opposition mounted by the Church.
  • 43.For an analysis of Berkeley’s phenomenalism, see G. Buchdahl, Metaphysics and the Philosophy of Science: The Classical Origins, Descartes to Kant (Oxford, 1969).
  • 44.For a penetrating and lucid critique of Newtonian empiricism, cf. P. K. Feyerabend, “Classical Empiricism”, in The Methodological Heritage of Newton, eds. R. E. Butts and J. W. Davis (Toronto, 1970). Here, Feyerabend dismantles the radical circularity that results from joining Newton’s philosophy of science with his actual procedure as a scientist. The two crucial moves that Feyerabend detects in Newton’s empiricism are (1) the identification of what Newton, in hisphilosophy of science, will call “phenomena” with his own actual experimental results, carefully selected and grossly idealised to conform to the theory that Newton hopes to “prove” (inductively, of course), and (2) the identification of these so-called “phenomena” (viz. his results) with “experience”. “It is therefore quite easy to turn part of the new theory into its own foundations by first presenting selected phenomena in its terms and by then pronouncing these phenomena to be the experience that has proved the theory ‘positively and directly’. Both identifications go almost unnoticed. Attacks upon the theory are soon answered by pointing out how firmly it rests on experimental fact. Attacks upon empiricism are answered by quoting the successes of the empirical rule of faith such as the theory of colours”. An earlier version of Feyerabend’s critique of classical empiricism can be found in “Problems of Empiricism, Part 1”,Beyond the Edge of Certainty: Essays in Contemporary Science and Philosophy, ed. R. Colodny (New Jersey, 1965). This phase of Feyerabend’s writings, 1965 to 1970, undoubtedly represents his best work, before the abrupt descent into methodological anarchism announced by “Against Method”,Minnesota Studies in the Philosophy of Science, vol. 4 (Minneapolis, 1970).
  • 45.For example, what does Marx mean when he writes, “Sense perception must be thebasis of all science?” That all our knowledge is grounded in “experience”, as classical empiricism maintained? In which case, what is experience? Or that the process of cognition starts with inductions and then proceeds by a more complex path? In which case, how does our knowledge acquire certainty? Or finally, that all our cognition regulates itself by reference to an “empirical base”? In which case, what is this “base”? In fact, it is pointless to take these questions seriously, simply because, in 1844, Marx endorsed this view on an ontological basis, with no regard for its implications at any other level. Later, of course, he wouldreject it completely, and make this very view a central part of his critique of vulgar economy. (e.g.,Theories of Surplus Value, Part One, p. 92, “coarse grabbing and interest in the empirically available material”).
  • 46.Bogdanov and Russian Machism are covered by D. Grille, Lenins Rivale: Bogdanov und seine Philosophie (Cologne 1966).
  • 47.Cf. Von Mises. “Ernst Mach and the Empiricist Conception of Science” in Mach: Philosopher,op. cit. and Frank,op. cit.
  • 48.Lenin, Materialism and Empirio-Criticism.
  • 49.Op. cit. p. 28.
  • 50.Op. cit. p. 56.
  • 51.But apodicticity is the hallmark of Lenin’s Materialism. It accounts for the circularity of so much of the argument there. A good example is the way he deals with Yushkevich, pp. 160ff.
  • 52. Heinrich Hertz, The Principles of Mechanics Presented in a New Form, tr. D. E. Jones and J. T. Walley (Dover, 1956), esp. the forty-page philosophical introduction which famously starts, “We form for ourselves images (innere Scheinbilder) or symbols (Symbole) of external objects; and the form we give them is such that the necessary consequents of the images in thought are always the images of the necessary consequents in nature of the things pictured” (p.1).
  • 53.Boltzmann, Theoretical Physics (n. 20 above). The young Wittgenstein’s picture theory of meaning was consciously modelled on the Hertz-Boltzmann model of scientific knowledge.
  • 54.Mach, cited Cohen, art. cit.
  • 55.Boltzmann, Theoretical Physics, p. 95–6.
  • 56.Boltzmann, Theoretical Physics, p. 225.
  • 57.Lenin, Materialism, p. 94.
  • 58.Lenin, Materialism, p. 169.
  • 59.Bacon, Novum Organum, Or True Suggestions for the Interpretation of Nature (1620),Aphorisms, Bk. 1, 50.
  • 60.Lenin, Materialism, p. 113.
  • 61.Op. cit., p. 114.
  • 62.Cf. G. W .F. Hegel, Logic, Part One of the Encyclopaedia, trans. W. Wallace (Oxford, 1975), p. 63.
  • 63.Lenin, Materialism, p. 117. As sensationalism here only means empiricism, this amounts to defining materialism as empiricism + a realist ontology.
  • 64.Colletti, Marxism and Hegel, p. 119.
  • 65.Lenin, Materialism, p.281. But Frank,op. cit. p. 9 uses Rey to argue that “Objective experience is not something which is outside and independent of our minds”.
  • 66.Hegel, Logic (Wallace), p. 67.
  • 67.Galvano Della Volpe, Logica come scienza storica (Rome, 1969), Ch. 2, where Della Volpe argues that Hegel, like Leibniz, simply “intellectualises” our sensations and consequently fails to establish a real basis for [fondare] the intellect itself beyond the “speculative notion” as a form accounting for its own contents. The element common to all the critiques referred to is the notion that our sensations form an indispensable element in our cognition and as such refer back to a “instance” external to thought itself, namely, to matter. For Della Volpe, the basic epistemological characteristic of this instance (of the material or the sensible) is its “discreteness” [discretezza]. A conceptual or rational equivalence of instances or particulars, or conceptual identity or unity, mediated through the discreteness of our sensations or of matter, establishes what Della Volpe callstautoheterology. A principle of non-contradiction, understood in a deeper epistemological sense, that is, by reference to the discreteness of matter, is thus essential to Della Volpe’s conception of “dialectic”.
  • 68.Della Volpe, op. cit. p. 155 ff.; adiluted form of Platonism because the latter is retained as support for the thesis that “instance and concept (parte e specie) are not the same thing”. Otherwise, of course, Della Volpe is as hostile to Plato as he is to Hegel.
  • 69.Many Galileos: for example, 1) the Galileo of Herschel, Mach, Heisenberg and the orthodox empiricist school. Herschel says, “Galileo refuted the Aristotelian dogmas respecting motion by direct appeal to the evidence of sense and by experiments of the most convincing kind” (cited Feyerabend, “Problems of empiricism”, art. cit.). But about how “convincing” Galileo’s experiments were see E. J. Dijksterhius,The Mechanization of the World Picture (Oxford 1961) and Koyré,op. cit. infra. For Mach’s Galileo, see D. Shapere,Galileo: A Philosophical Study (Chicago 1974). For Heisenberg’s Galileo, cf. Werner Heisenberg,Das Naturbild der heutigen Physik (Hamburg 1965) p. 60 and the idea that Galileo first enunciated “das Grundprinzip, das die Wechselbeziehung zwischen Hypothesen und Erfahrung festlegt”; this is argued on the basis of Galileo’s letter to Carcarilla, 5 June 1639, where Galileo in fact says the reverse of what Heisenberg takes him to be saying. 2) The empirico-rationalist Galileo of Geymonat (op. cit. n. 112 below), who balances between “rigorous demonstration” and “sensory experience”. This is likewise Della Volpe’s Galileo, though, in Della Volpe, he inclines more towards empiricism. 3) The more discernibly classical-rationalist Galileo of Ernst Cassirer, anticipating Kant’s own “critical” concept of experience. 4) The consciously Platonising, thoroughly mathematicist, purely rationalist Galileo of Alexandre Koyré,op. cit. infra.  And 5) the counter-inductivising Galileo of Feyerabend,Against Method (London 1975).
  • 70.Cf. Feuerbach, “Thinking posits the discreteness of reality as a continuum, the infinite multiplicity of life as an identical singularity”, cited Lenin, Philosophical Notebooks, p. 83; and see Della Volpe,Logica, Ch. 4, for discreteness or the punctual nature of matter.
  • 71.Dell Volpe, Logica, pp. 179-80; an open, conscious, validation, by a Marxist, of eliminative inductivism.
  • 72.H. W. B. Joseph, An Introduction to Logic (Oxford, 1925), cited Della Volpe, p. 171.
  • 73.Cf. Della Volpe, Logica, p. 184 for the “corrigibility” of scientific laws, p. 186 for the importance of “prediction”, pp. 170-71 for experimental verifictions that turn “hypotheses” into “laws”. For Della Volpe, as for Popper, predictivism has more than a merely methodological significance, it is a rationality-criterion, an expression of the “honesty” that characterises science against metaphysical dogmatism.
  • 74.Logica, p. 205, which refers to “il contraddistinguersi del marxismo come metodo non solo dall’ idealismo e le sue ipostasi ma altresi dal positivismo con la sua idolatria dei ‘fatti’” and positivism’s “ripugananza alle ‘ipotesi’”. Up to 1956, in fact, both editions of Della Volpe’s work were entitled Logica come scienza positiva.Positiva was subsequently changed tohistorica. The book itself contains a nuanced and not entirely disapproving assessment of logical positivism (pp. 251ff). Here again, the main thrust of the critique is directed against the purely conventionalist elements in positivist thought (with Della Volpe citing Lenin on Mach).
  • 75.Lukács, History and Class Conciousness (London, 1971), p. 5.
  • 76.Della Volpe, Logica, p. 175 for “scientific” or “material” ontology as against “formal” or “metaphysical” ontology.
  • 77.Schmidt’s book is undoubtedly valuable for its many correct insights into the theme he sets himself. He is, for example, much more sensitive to the classical German heritage in Marx. But the book as it stands is quite confused and this basically because of the assumption it makes that Marx evolved all in one piece. A significant index of this is his omission of any discussion of Marx’s pre-Feuerbachian views, for example, those elaborated in and around Marx’s doctoral dissertation. The discussion in Schmidt starts with a Feuerbachian Marx, and because it assumes perfect continuity, his own exposition is fraught with the tensions of Marx’s development.
  • 78.Op. cit., p. 29. As argued earlier, sensualism is strictly an ontological empiricism with an only implicit epistemological (or sensationalist ) content.
  • 79.Schmidt, Concept of Nature, pp. 115–16.
  • 80.Op. cit, p. 116.
  • 81.Op. cit., p. 115f. There is nothing wrong with rendering “epistemology” genetic, but it cannot be done in the way Schmidt suggests.
  • 82.Op. cit., p. 121–22.
  • 83.Cf. Lukács, “Moses Hess”, Op. cit., p. 204, note 47, citing Marx’s letter to Engels, 24 April 1867.
  • 84.Marx, Early Writings, p. 389; emphasis mine. (All references in this section are to theManuscripts.)
  • 85.Cf. Sartre, Critique de la raison dialectique (Paris, 1960), pp. 18 ff. Soren Kierkegaard, active in the very decade that Feuerbach and Marx turned to the critique of Hegel, devoted a lot of his work to a refutation of Hegel. Much later, Jean Wahl,Le malheur de la conscience dans la philosophie de Hegel (Paris, 1929) would attempt a sort of reconciliation, by Kierkegaardianising thePhenomenology. The respective fates of Feuerbach and Kierkegaard are an interesting index of the climate of European thought in the closing decades of the 19th century and the early part of this century. Kierkegaard left an indelible imprint through existentialism (for the importance ascribed to him, cf. the Gallimard collectionKierkegaard vivant, Paris 1966, with essays by Sartre, Marcel, Heidegger, Paci, Jaspers, and Wahl). Feuerbach, by contrast, was rapidly forgotten and has come back into attention purely as a function of the revived interest in Marx’s early writings. The difference lies, of course, in the fact that Feuerbach advocated an ontology, Kierkegaard aLebensphilosophie. By the 1870s, Europe was no longer interested in metaphysical refutations of metaphysics. It found its typical symbols in Mach on one side, Nietzsche on the other. Between them, Mach and Nietzsche exhaust the whole trajectory of the European intelligentsia over the next forty years.
  • 86.Early Writings, pp. 396, 393.
  • 87.Hegel, Logic,Op. cit. p. 175. Over against this Hegelian notion of what Della Volpe will later call tautoheterology stands the “abstract identity” of formal logic. Abstract identity = tautology.
  • 88.Marx, Early Writings, p. 388.
  • 89.Early Writings, p. 398.
  • 90.Early Writings, p. 399.
  • 91.Early Writings, p. 399.
  • 92.Early Writings, p. 400.
  • 93.Early Writings, pp. 349, 355.
  • 94.Early Writings, p. 355, “Only when science starts out from sense perception in the dual form of sensuous consciousness andsensuous need…”.
  • 95.Early Writings, p. 389.
  • 96.Early Writings, p. 390.
  • 97.Early Writings, p. 390.
  • 98.Early Writings, p. 391. D. McLellan, Marx before Marxism (London 1970), p. 200 in fact says that Marx “later crossed out” the very sections I have been talking about.
  • 99.Early Writings, p. 391.
  • 100.Early Writings, p. 328-29.
  • 101.Capital, vol.1 , tr. Ben Fowkes (London 1976), p. 284.
  • 102.Schmidt, Concept of Nature in Marx, p.27. All physicalism must end with an operationalist theory of meaning. And that exactly is what Schmidt does when he writes, “We only really know what a natural thing is when we are familiar with all the industrial and experimental-scientific arrangements which permits its creation” (p. 122), with which compare the identical conception of the positivist P.W. Bridgman,The Logic of Modern Physics (New York, 1927). For Bridgman, only those concepts have any “meaning” in science that can be understood in terms of the experimental operations that they allow, or prescribe, or enable.
  • 103.From the preface to the English edition of The Concept of Nature in Marx, p. 10, referring to a 1967 essay in German called “For a New Reading of Feuerbach”. The swing back to Feuerbach was predetermined by the confusions latent in Schmidt’s argument.
  • 104.Op. cit., p. 27–8.
  • 105.On the other hand, when they return to Feuerbach in The German Ideology (London, 1965) pp. 57-59, Marx and Engels revert to the sort of interpretation Schmidt proposes. They contrapose sensuousness as contemplation to sensuousness as physical (sensuous) activity. Thus, these pages simply fail to draw out the content of the distinction that Marx establishes in the first thesis. Again, it is beside the point to see in this retrogression the hand of Engels.
  • 106. Early Writings, pp. 391, 389 and passim. With all of which cf. the terminology of the rationalist psychology of the sixteenth and seventeenth centuries, e.g., Juan Huarte’s “generative powers”, Leibniz’s “innate dispositions”, Spinoza’s “native powers of the intellect”. With “drive” cf. Aristotle’shorme (Ὁρμή),  which he sees implanted in the whole of nature. The Aristotelianhorme or drive manifests itself asorexis (desire).
  • 107.Early Writings, pp. 386, 395, 357.
  • 108.Early Writings, p. 352–353.
  • 109.But earlier than Kant, in the whole Rationalist tradition, of course, and before that in Renaissance humanism.
  • 110.Marx, Early Writings, p. 354; I have italicised “psychology”.
  • 111.Marx, Early Writings, p. 389; italics mine.
  • 112.Holy Family, p. 177.
  • 113.Cf. Pico’s claim, “The very nature of sense is various, not only because of variations in the object of sense, but because of variations in the human constitution, which is also various in its own nature”, in Examen vanitatis doctrinae gentium (1520). Thus Ernst Cassirer,Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeit, Bd. 1 (Berlin, 1911) distinguishes this early Platonism “from the truly modern form of Platonism which grew up on the basis of the exact sciences”. Ludovico Geymonat,Galileo Galilei, tr. Stillman Drake (New York, 1965; Italian orig. 1957) makes basically the same point: “The Platonic legacy of Marsilio Ficino had passed to the philosophers of southern Italy, and not into the cultural heritage of Florence, which was dominated by the practical spirit of Machiavelli, and it was there that Galileo grew up” (p. 33).
  • 114.See N. Chomsky, Language and Mind (New York 1968), pp. 70 ff. Also see Chomsky’s note 11 at p. 83 for Kant and Leibniz. For Descartes, cf. Blakeet al.,Theories of Scientific Method, pp. 99f. “However much Descartes may speak of ‘innate ideas’ … the fact nevertheless remains that for Descartes, as for Aristotle, the process by which these ‘innate ideas’ are actually brought into explicit consciousness is one which as a matter of fact begins with experience of particulars”. (I shall come to Hegel later.) Colletti’s formulation of the problem is thus not only wrong in accepting a crude version of innate ideas as the only one, but also incomprehensible in the terms in which it is put: “The alternative is simple: either one assumes that the real objects to be known are given, or else it has to be that the known ‘object’ is already givenqua knowledge itself, as ‘innate’ knowledge” (Marxism and Hegel, p. 86) . The latter part of this sentence is simply devoid of sense.
  • 115.Leibniz in his letter to Queen Sophia Charlotte of Prussia, 1702, reprinted in M. Hollis, The Light of Reason. Rationalist Philosophers of the Seventeenth Century (London 1973), p. 333.
  • 116.Kant, Anthropology  from a Pragmatic  Point of View, p. 7; cf. alsoCritique of Pure Reason, p. 55, “it must be admitted that the Leibnizian-Wolffian philosophy has assigned an entirely erroneous point of view to all investigations into the nature and origin of our cognitions, in as much as it regards the distinction between the sensuous and the intellectual as merely logical, whereas it is plainly transcendental, and concerns not merely the clearness or obscurity, but the content and origin of both”.
  • 117.Cf. Galileo, The Dialogue Concerning the Two Chief World Systems, tr. S. Drake (Berkeley 1953), day 1, where he says, “Taking man’s understanding intensively, insofar as this term denotes understanding some proposition perfectly, I say that the human intellect does understand some of them perfectly , and thus in these it has as much absolute certainty as nature itself … its knowledge equals the divine in objective certainty, for here it succeeds in understanding necessity.”
  • 118.G. Lukács, Geschichte und Klassenbewußtsein (Berlin, 1968), pp. 217-220; Lukács, History and Class Consciousness, pp. 117f. But as the example of Descartes shows, geometricism was often a purely expository device, a way of ordering an argument more rigorously, cf. Descartes’ Geometrical Appendix to theSecond Replies (1641). Descartes would repeatedly make the point that “the complete mathematization of physics is impossible” for “as soon as mathematics ceases to be pure. They require a datum to which they are applied and which they interpret, but which they accept without themselves being able to justify”. See also the fifth rule in hisRegulae ad directionem ingenii where he refers disparagingly to “those philosophers who proceed to the neglect of experience (neglectis experimentis) and imagine that truth will spring form their brain like Pallas from the head of Zeus”, Descartes,Philosophical Writings, tr. Norman Kemp Smith (London 1952), p. 23.
  • 119.B. Spinoza, On the Improvement of the Understanding (1661), tr. J. Katz (New York ,1958) pp. 12, 35, 25 andpassim. Cf. also Katz’s introduction, p. xiv, “How Spinoza thinks that truth makes itself manifest … is still vigorously debated… He seems to think of truth as constituted by proper deduction from premises, or as making a particular statement coherent with a system of statements, or even as building up our statements by starting with definitions”. It is interesting that Marx very early on in his youth rejected precisely the “mathematical” dogmatism of his own earlier metaphysic of law, cf. the letter to his father dated 10 November 1837: “First came what I was pleased to christen the metaphysics of law, that is, foundational propositions, reflections, and conceptual determinations that were separated from all actual law and from every actual form of law, just like in Fichte, only in my case it was more modern and less substantial. Moreover, the unscientific form of mathematical dogmatism – where the subject runs around the matter, here and there rationalizing, while the topic itself is never formulated as a richly unfolding living thing – was from the very beginning a hindrance to grasping the truth. The triangle allows the mathematician to construct and to demonstrate, yet it remains a mere idea in space and doesn’t develop any further… By contrast, in the concrete expression of a living concept world, as in law, the state, nature, and all of philosophy, the object must be studied in its development … and the reason of the thing itself must be disclosed as something imbued with contradictions and must find in itself its unity” (tr. Paul Schafer,The First Writings of Karl Marx, p. 74). Jean Hyppolite,Hegel. Préface de la Phénoménologie de l’Esprit (Paris 1966), p. 202 notes that for Hegel the opposition between mathematical truth and philosophical  truth was a fundamental one. See Hegel’s own contrast betweenhistorische und mathematische Wahrheit  in the Preface to the Phenomenology (Op. cit., pp. 96ff) and the implied criticism that “die Bewegung des mathematischen Beweises gehört nicht dem an, was Gegenstand ist, sondern ist ein der Sache äußerliches Tun”, basically the same point that Marx makes, of course. All this was before the peculiar dissociation between “proof” and “truth” within mathematics itself that was brought about by Gödel’s incompleteness theorems (1931).
  • 120.Early Writings, p. 392.
  • 121.Feuerbach, cited Lenin, Notebooks, p. 67: “unableitbares, ursprungliches Wesen”, about nature.
  • 122.This has been the traditional interpretation of Hobbes; cf. De Corpore (1655), chapter 6, on method (cited Hollis ,Op. cit., p. 178ff.). Sections 13ff. of chapter six deal with “definitions”. In their discussion of the chapter Blakeet al.,Op. cit. ask how Hobbes’s “principles” are themselves established if not through induction. But with Hobbes’s whole conception both of “definitions” and of the analytic method, compare both Bacon’s notion ofform (Novum Organum, 2.13, “the form of a thing is the very thing itself, and the thing differs from the form no otherwise than as the apparent differs from the real, or the external from the internal”, and 2.20, where form or “true definition”  “consists in this, that it furnishes an idea of  the true thing”) as well as Marx’s conception of form (e.g.form of value ). The analysis-synthesis conception that Hobbes takes over from Galileo goes back earlier than Galileo to the logical thought of the Renaissance.
  • 123.Feyerabend, “Classical empiricism”, p. 151. [Reprinted in P. K. Feyerabend, Problems of Empiricism (Philosophical Papers Vol. 2) (Cambridge, 1981), where this passage is at p. 35.] Both Feyerabend and Koyré are emphatic about Aristotle’s empiricism, though the former agrees that the theory of “powers” and “activities” transgresses the limits of empiricism (art. cit., p. 232, note 31). On the other side, Hegel,Lectures on the History of Philosophy, is emphatic in the opposite view, rebutting the notion of Aristotle’s empiricism in some extremely careful analyses of theDe Anima. For Della Volpe, Aristotle’s greatness is his “materialism”, taken to mean specifically the transcendental use of the principle of non-contradiction (Op. cit. p.149), in other words,  the material substratum is itself or has its roots in – “nell’atto del percepire, cioe nella conoscenza ch’e gia la sensazione in atto”.
  • 124.For the successive drafts of this rule, see Koyré, “Newton’s Regulae philosophandi”, republished inNewtonian Studies (Chicago 1968), pp. 269ff. Here, Koyré subjects the whole Newtonian conception of “hypotheses’’ to painstaking analysis, and suggests that Newton’s polemic is directed against Descartes.
  • 125.See the analogy, a superb one, that Feyerabend draws between Lutheran fideism and classical empiricism, art . cit.
  • 126.One “solution” to all this is the notion that all our cognitive claims about the world are purely “conjectural”. For example, Karl Popper, Conjectures and Refutations (London 1969), chapter one, where he says, “all laws, all theories, remain essentially conjectural or hypothetical”. On the other hand, Engels,Dialectics of Nature (Moscow 1974) pp. 240, 232, proposes a more complex relationship between “laws” and “hypotheses”. He sees the notion of “laws” in a traditional sense (pre-conventionalist), but the process of their discovery as one of the continuous proposal of “hypotheses”. Unfortunately, this conception rests on the crude distinction that Engels draws between observations (or “facts”) and theories (or explanations). Della Volpe inclines Engels in Popper’s direction.
  • 127.Hegel, Logic, p. 63.
  • 128.Sartre, “Materialism and Revolution”, in Literary and Philosophical Essays (London 1968), p. 188: “Materialism in a metaphysics hiding beyond positivism”.
  • 129.Op. cit. p. 28, nullifying Schmidt’s whole concept of “mediation”.
  • 130.Timpanaro, On Materialism, pp. 80, 34. I think Sartre’sCritique fails to avoid the tension between tautology and heterology. Like Marx in theManuscripts, he tries to overcome it through a dialectic of intersubjectivity, hence the crucial role, both in theManuscripts and in theCritique, ofbesoin, in Sartre the practico-inertmotor of the dialectic.
  • 131.Hegel, Logic,loc. cit., “So long, then, as this sensible sphere is and continues to be for empiricism a mere datum, we have a doctrine of bondage”. Timpanaro is right – “epistemology has been charged with the task of founding the absolute liberty of man” (p. 35, and n. 8 at p. 81f.).
  • 132.Hegel, Lectures on the History of Philosophy , tr. E. S. Haldane and F. Simon (London, 1968), vol. 1, p. 20, from the Introduction.
  • 133.J.H. Randall, Aristotle (New York, 1960), p. 170.
  • 134.Randall, Op. cit. pp.65, 129f., summarisingMetaphysics, Book Theta.
  • 135.W. Von Humboldt, Über die Verschiedenheit des menschlichen Sprachbaues (Berlin 1836), cited Chomsky,Current Issues in Linguistic Theory (The Hague, 1969) pp. 17f.
  • 136.Chomsky’s most explicitly philosophical discussion of this is in Cartesian Linguistics: A Chapter in the History of Rationalist Thought (New York, 1966) andLanguage and Mind, Ch. 3. The identification of “generative grammar” withcompetence arises only in Chomsky’s later work, where it is first signalled byAspects of the Theory of Syntax (Cambridge, Mass., 1965). Chomsky’s whole work arose as a polemic against the empiricist (or behaviourist) accounts of language-acquisition represented notably by Bloomfield and Skinner. Structuralising discussions of Chomsky, e.g., P. Petit,The Concept of Structuralism: A Critical Analysis (Dublin 1975) show simply no interest at all in the cognitive foundations and implications of Chomskyan linguistics. Colletti’s hostility to the innateness “hypothesis” has already been alluded to. For Timpanaro, cf.On Materialism, p. 199f, and p. 203, n. 147, “the merits of empiricism and of its anti-innatist polemic”. If any sort of psychology corresponds to Timpanaro’s materialism, it is obviously something close to behaviourism.
  • 137.Hegel, Logic, p. 224f. Cf. Chomsky,Language and Mind, p. 67: “The Platonistic element in Humboldt’s thought is a pervasive one; for Humboldt it was as natural to propose an essentially Platonistic theory of ‘learning’ as it was for Rousseau to found his critique of repressive social institutions on a conception of human freedom that derives from strictly Cartesian assumptions regarding the limitations of mechanical explanation”.
  • 138.Hegel, Logic, p. 261.
  • 139.On Mechanism and Teleology, see Hegel, Science of Logic, trans. A. V. Miller (London 1969), pp. 711ff., 734ff.; cf. the smallerLogic, pp. 261ff. For the historical background to the rise of a “mechanistic” world picture cf. Dijksterhuis,Op. cit. and R. S. Schofield,Mechanism and Materialism: British Natural Philosophy in an Age of Reason (Princeton, 1970).
  • 140.Sartre, Being and Nothingness. An Essay on Phenomenological Ontology (London, 1966), p. 613.
  • 141.Sartre, Being and Nothingness, pp.614.
  • 142.Hegel, Logic  (Wallace), §204 (p. 267).
  • 143.Being and Nothingness, p.615. Again the hole polemical thrust of these pages lies against behaviourism.
  • 144.Hegel, Logic, §234 (p. 291).
  • 145.Logic, p. 290.
  • 146.Logic, §206 (p. 270).
  • 147.Marx, Capital, Vol. 1, p. 284. The theme is known as the “teleology of labour” and discussed in Lukács, The Young Hegel (London 1975), pp. 338ff.
  • 148.Hegel, cited Lukács, Young Hegel, p. 363.
  • 149.Hegel, Logic (Wallace), p. 268; “the End” modified to “humans”. For the “cunning” of reason,ibid., p. 272.
  • 150.Hegel, Science of Logic, p. 737.
  • 151.Marx, Early Writings, p. 398.
  • 152.Hegel, Logic (Wallace), §145 (p. 206), where he also says “the Freedom of Will is an expression that often means mere free choice, or the will in the form of contingency”.
  • 153.H. Holz, Leibniz (Stuttgart, 1958), p. 20f. where he writes that occasionalism, French materialism, Spinoza and Leibniz each in their own way attempted to overcome the dualism of Cartesian ontology. “Das Zeitalter Leibniz steht, philosophiegeschichtlich gesehen, unter dem Zeichen der Auseinandersetzung mit dem cartesischen System. So ist auch das Leibniz’sche Denken metaphysisch als Gegenzug gegen den Cartesianismus entstanden. Der cartesische Dualismus vonres cogitans undres extensa hatte die Begrifflicheinheit der Welt auseinandergerissen”. Holz subsequently makes an extremely artificial attempt to reconstruct the Leibnizian monadology as a prefiguration of dialectical materialism.
  • 154.Reported in Aristotle, Metaphysics, I, 3.
  • 155.Hegel, Lectures, vol. 1 p. 348, citing Sextus Empiricus,Adversus Mathematicos, 7, 89. The same view is ascribed by Sextus to Heraclitus. In Antiquity the radically opposite view, upholding the veracity of our sensations, was most forcefully argued by Epicurus. The main evidence for this ancient empiricism lies in Diogenes Laertius, who says, “Now in the Canon Epicurus affirms that our sensations … and our feelings are the standards of truth”. Reason cannot “refute them, for reason is dependent on sensation”. “The reality of … perceptions guarantees the truth of our senses”. “It is from phenomena that we must seek to obtain information about the unknown. For all our notions are derived from perceptions, either by actual contact or by analogy or resemblance, or composition, with some slight aid from reasoning”,Collected Works, vol.1, p.405f.). Anaxagoras’ fragments have been collected together inAnaxagoras. Testimonianze e frammenti , ed. De Lanza (Rome, 1966), Epicurus’s, inEpicuro: Opere,  tr. G. Arrighetti (Turin, 1960).
  • 156.Hegel, Lectures on the History of Philosophy, vol. 1, p. 373; cf. Marx,Notebooks on Epicurean Philosophy, inCollected Works, vol. 1, p. 436, “with the Sophists … it is ideality itself which, in its immediate form, thesubjective spirit, becomes the principle of philosophy”, and the remarks that follow this in the Second Notebook.
  • 157.Hegel, Lectures, p. 374.
  • 158.Hegel, Lectures, p. 385.
  • 159.Hegel, Lectures, p. 386. Again, cf. Marx,Notebooks, p.439, “Subjectivity is manifested in its immediate bearer [Socrates] as his life and his practical activity…apart from this practical activity, his philosophy has no other content than the abstract determination of the Good”.
  • 160.Plato, Phaedo, ed. G. Stallbaum (Leipzig, 1866), p. 72e,hemin he mathesis ouk allo ti e anamnesis tynchanei ousa. The connection I have drawn between Plato and the earlier tradition is stated as follows by Marx,Notebooks, p.490f.: “As the nous of Anaxagoras comes into motion in the Sophists … and this immediate daemonic motion as such becomes objective in the daemon of Socrates, so alsothe practical motion in Socrates becomes a general and ideal one in Plato, and the nous expands itself into a realm of ideas”. Likewise, in the 2nd Notebook, “With Plato motion becomes ideal; as Socrates is the image and teacher of the world, so Plato’s ideas, his philosophical abstraction, are its prototypes. In Plato this abstract determination of the good, of the purpose, develops into a comprehensive world-embracing philosophy” (p. 439). But the interpretation that then follows of the Theory of Ideas is a fairly traditional one. The term “purpose” in Marx’s text refers to the notion ofaition in Socrates. Thus in thePhaedo (Op. cit., p.99b) Socrates will criticise Anaxagoras for confusing the “true purposes” or “ends” of human activity (theaitia) with those conditions that render the execution of such purposes possible (e.g., the fact that humans have bodies): “It shows an inability to distinguish between that without which this real cause would not be a cause at all (aneu hou to aition ouk an pot’aie aition)”. (This is also cited by Hegel in hisLectures, p. 342.)
  • 161.Hegel, Lectures, vol.2, p.32f.; cf. Humboldt, “Die Erlernung ist…immer nur Wiedererzeugung”.
  • 162.Hegel, Lectures, p. 43. So we must add Hegel to a list that includes Kant, Leibniz, and Descartes. To repeat, none of the major proponents of the theory of innate ideas ever denied the role of “experience” or “perceptions”, etc., in the manifestation of those “ideas”.
  • 163.Hegel, Lectures, p. 45. Note well, “the source of knowledge in manifold”.
  • 164.Cited Hegel, Lectures, p.64. The basic study of Plato’s dialectic remains J. Stenzl,Studien zur Entwicklung der platonischen Dialektik von Sokrates zu Aristoteles (Leipzig, 1931).
  • 165.Marx, “Die moralisierende Kritik und die kritische Moral”, cited Schmidt, Op. cit., p. 50.
  • 166.Hegel, Lectures, p. 140. On the other hand, W. Heisenberg,Physik und Philosophie (Verlag Ullstein, 1959), p. 44 writes that modern physics vindicates Heraclitus more than anyone else.
  • 167.Hegel, Lectures, p. 141.
  • 168.Hegel, Lectures, p. 203. Hegel also says, “In the Kantian philosophy we for the first time have that (sc. Aristotle’s) conception once more awakened in us … life has there been made an end to itself” (p. 160).
  • 169.I. Kant, Groundwork of the Metaphysic of Morals, in H.J. Paton,The Moral Law (London 1966). Without this Aristotelian-Kantian ethical rationalism, the Marxist tradition will simply continue to suffer from pure ethical pragmatism. Marx himself unreservedly accepted the notion of human nature “as an end in itself”.
  • 170.Marx, Capital, vol.3 (Moscow 1959), p. 800.
  • 171.Engels, Dialectics of Nature, pp. 21-22. The passage is unparalleled for the admiration it expresses, and the sheer enthusiasm it shows, for an earlier period of European thought. Generally, both Marx and Engels are severe in their judgements about most things. How often do you find words like “giants” in their writing?
  • 172.Cf. Geymonat, Galileo, p. 71ff.
  • 173.All citations taken from Blake et al., Theories of Scientific Method, pp. 11ff., referring mainly to hisFrammenti letterari e filosofici (Florence, 1925).
  • 174.Thus Cassirer, Das Erkenntnisproblem, says about da Vinci, “So gründet Leonardo seinem Idealbegriff der Wahrheit und der Vernunft in dem fruchtbaren ‘Bathos der Erfahrung’, während umgekehrt der Erfahrungsbegriff selbst seinem Wert aus dem notwendigen Zusammenhang erhält, in dem er mit der Mathematik steht… Die Erfahrung selbst ist nichts anderes als die aussere Erscheinungsform der Vernunftbeziehungen und Vernunftgesetze”.
  • 175.Koyré, Études galiléennes, 3 vols. (Paris 1939), cited Geymonat,Op. cit., p. 178.
  • 176.Alexandre Koyré, “Galileo and Plato”, in Metaphysics and Measurement: Essays in Scientific Revolution (London 1968), p. 38. Cf. “Aristotelean physics is based on sense-perception, and is therefore decidedly non-mathematical. It refuses to substitute mathematical abstractions for the colourful, qualitatively determined facts of common experience, and it denies the very possibility of a mathematical physics on the ground (a) of the non-conformity of mathematical concepts to the data of sense-experience, (b) of the inability of mathematics to explain quality and to deduce movement” (p. 5). This opposition of “quality” and “mathematics” forms a basicleitmotif of Koyré’s entire Galileo interpretation. Koyré was a student of Husserl, but was influenced, presumably (if at all!), by the early Husserl of theLogische Untersuchungen, with his sharp attack on “psychologism”.
  • 177.Geymonat, Galileo, pp. 179, 13, 181. Geymonat discusses both theDialogue and theTwo New Sciences in some detail. While the whole book can be seen as a mild polemic against Koyré, Geymonat’s own formation being more strictly positivist, he does not fully accept Della Volpe’s Galileo either. For example, he writes, “Della Volpe appeals to the decisive role which Galileo attributed to experimental verification, andleaves aside the problem of interpreting the role he attributed to mathematics” in Galilean science (p. 185). And note the tone of caution in passages such as these: “Why not suppose that Galileo regards mathematics more in its technical aspect, as an aid to logic, than metaphysically, as the expression of a more stable and harmonious reality underlying the fluctuations of phenomena?” (p. 108). Or, “Perhaps Galileo wanted only to prove to everybody that mathematical demonstration even though it absorbs into itself the experimental, is something independent of experiment, and that the two are developed on entirely separate planes” (p. 181).
  • 178.Koyré, Op. cit. p. 90.
  • 179.Claude Bernard’s major work, Introduction à l’étude de la medicine expérimentale (1865), was translated asAn Introduction to the Study of Experimental Medicine (New York 1957). Despite its Althusserian undertones, Hirst provides a good short account of Bernard’s views on experiment inEconomy & Society, 2 (1973). Della Volpe,Logica, p. 173 was likewise impressed by Bernard’s capacity to avoid both apriorism and empiricism.
  • 180.Koyré, Op. cit., p. 90.
  • 181.Geymonat, Galileo, p. 51.
  • 182.Lenin, Materialism, pp. 123ff., 56.
  • 183.Lenin, Materialism, pp. 129-30; italics mine.
  • 184.Della Volpe, Logica, p. 169f., and Appendix 1.
  • 185.Logica, p.201; italics mine. The passage reads, “Cosi le fondamentali ipotesi marxiane del valore-lavoro et del connesso plusvalore, etc., tali ipotesi fondamentali, hanno acquistato la verità-realtà di leggi quando la practica esperienza storico-economica del capitalismo di monopolio degli ultimi cinquant’anni ha confermato che un fenomeno grave come quello della crisi, ad esempio, non si puo spiegare che con la contraddizione organica dell’economia di profitto capitalistica … contradizione prevista … nella ipotizzata connessione basilare di valore-lavoro e plusvalore appunto”; then going on to cite the lines from Lenin I quoted above (p. 87). The idea of relating the “truth” of Capital to the growth of monopoly capitalism was probably suggested to Della Volpe by Maurice Dobb, whom he quotes on p. 202. Mandel is another classic example of the “predictive power” argument, e.g. in his introduction to Ben Fowkes’s translation ofCapital, vol.1, where he replies to Popper as follows: “This is obviously based upon a misunderstanding of the very nature of the materialist dialectic, which, as Lenin pointed out, requires constant verification through praxis to increase its cognitive content. In fact, it would be very easy to ‘prove’ Marx’s analysis to have been wrong, ifexperience had shown for example that the more capitalist industry develops, the smaller and smaller the average factory becomes” (p. 24), and so on.
  • 186.Logica, p. 202 “del criterio della practica che convalida ossia verifica la ipotesi tramutandalo in legge”.
  • 187.Logica, p. 186, “Onde si debba sostituire in ogni campo laprevisione scientific (ipotesi)”. This is a thoroughly Popperian sentiment, almost down to the last word, except that Popper himself is more consistent. He asks, whatform must our cognitive claims take to count as “predictions”? See the  revealing autobiographical details inConjectures and Refutations, Ch. 1.
  • 188.Logica, p.189 “che tali ipotesi valutativa sia verificabile practimente ossia attraverso l’esperimento storico…”
  • 189.Logica, p.188 “la practica umano-sociale oesperimento storico”.
  • 190.Marx to Kugelmann, 11 July, 1868 (Selected Correspondence, p. 251).
  • 191.Hegel, Logic (Wallace), p. 110.
  • 192.Galileo, Il Saggiatore, cited Koyré,Op. cit. p.34, note.
  • 193.Capital, vol.1, p.433.
  • 194.Galileo, cited Feyerabend, Against Method, p. 71. Feyerabend develops this line of argument into the notion of the “mutability” of all “experience”,  or its “fluidity”;  cf. p. 89, “Experience now ceases to be the unchangeable fundament that it is both in common-sense and in the Aristotelian philosophy”. The notion of the “fluidity” of experience is then likewise one that underlies Kuhn’s model of the paradigm-dependence of our “perceptions”,The Structure of Scientific Revolutions (Chicago 1968). Kuhn’s thesis was a fusion of (a) the central role that Koyré ascribed to ontological shifts in any scientific revolution and (b) Gestalt theories of perception.
  • 195.Galileo, Dialogue, day 3 (tr. Drake, p. 328).
  • 196.All this alludes of course to the most radical proponent of this conception, Gaston Bachelard, who completely rejects the notion of any sort of “givenness” in science, e.g. in L’activité rationaliste de la physique contemporaine (Paris, 1951), p. 87, “unless we choose completely to alter the meaning of words, we cannot really say that the corpuscules with which contemporary science deals are givens (donneés). They are certainly not givens of experience, but nor are they really given at a hidden level (donneéscachés). They are established by invention, not by discovery (Il faut plutôt les inventer que les découvrir)”. Earlier, inLa dialectique de la durée (1936): “We have reached a level of knowledge at which scientific objects are what we make them, no more or no less…”. Experiments to Bachelard imply techniques that are themselves only technical materialisations of “reason”, so that their “positive results” in no sense entail a rehabilitation of “the absolute positivity of experience as such”. InLe Rationalisme appliqué (Paris 1962), he makes the point as follows: “Technical materialism is absolutely not a philosophical realism. It corresponds basically to a transformed reality, or an adjusted reality, or a reality that bears the imprint of human intervention, or a sign of reason”. Thus the “reality” of cognitive reason is one that changes constantly (cf. Feyeraband, n. 194 above). E.g. “the ‘electrical reality’ of the 19th century is quite different from the ‘electrical reality’ of the 18th.” One of the better known arguments for this is Bachelard’s analysis of the successive notions of “substance” inLa philosophie du non (Paris, 1966), where Bachelard remarks that the “realism of substance” that permeates classical chemistry holds true only as a very crude first approximation (Op. cit., p. 70ff.), as a sort of blind, spontaneous philosophy of science.

      The influence of Bachelard on Althusser has been grossly exaggerated. If there is any definite philosophical ancestry to Althusser’s conception of “Science”, then this is more easily traceable to Spinoza and the identification of mathematical truth as the pure model of all cognitive truth (cf. n. 119, p. 56–7 above). What Althusser encounters inCapital is really the dialectic in its Hegelian sense. But Althusser’s positivism, his sustained polemic against the whole of classical philosophy (Spinoza excepted) precludes a “Hegelian” reading ofCapital. The solution is then found in a systematic reconstruction of the dialectic in the image of a Spinozist-mathematicist deductivism. Now Lukács had already noted, in a profound comment (History and Class Consciousness, p. 143), that “history was an insuperable barrier to the (classical) Rationalist theory of Knowledge”. Thewhole tendency of classical German philosophy following Kant was precisely towards the resolution of this “problem” through the reconstitution of rationalist epistemology on a dialectical terrain. Thus one inevitable consequence of Althusser’s positivistic retrogression to classical rationalism is therenovation of this problem in its more or less classic form. Recall Lukács again (cited p. 55 above) that “the principle of systematization is not reconcilable with the recognition of any ‘facticity’”. This is where Bachelard comes into Althusserianism, for precisely Bachelard’s central thesis is that within cognitive reason thereis no such “facticity”, no givenness, but that all scientific objects are constructed objects. Althusser does not go all the way, however, and this is what accounts for the rapid disintegration of his school into those who, like Balibar, veer back to “history”, and others, in England, who deny that there can be a “science of history” at all. Althusser, caught in between, remains diplomatic and centrist.

  • 197.Marx, Early Writings, p. 398.
  • 198.Cited Koyré, Op. cit., p. 13.
  • 199.Dialogue Concerning Two New Sciences, cited Geymonat, Op. cit., p. 179.
  • 200.Da Vinci, Frammenti,Op. cit., p. 87.
  • 201.The initial move comes with conventionalism. In Materialism, p. 297ff. Lenin sees this general drift away from justification as ahealthy move towards the thesis of the “relativity” of all our cognitions, but argues that Duhem et al., ignore “dialectics”, they collapse this relativism into idealism.
  • 202.Leibniz, Philosophical Writings, p. 182.
  • 203.This is a central point in Feyerabend’s critique of empiricism, e.g. “How to be a Good Empiricist”, in Philosophy of Science: The Delaware Seminar, ed. B. Baumrin (New York, 1963), reprinted elsewhere several times over due presumably to its powerful undercurrent of liberalism. The cue is taken from K. Popper,The Logic of Scientific Discovery (London, 1974), Ch. 5 which is called “The Problem of the Empirical Basis”. Here, Popper proposes the notion of “basic statements” [Basissatze] that take the form of “singular existential statements” as a solution to the problem of the “objectivity” of the empirical basis. But already here, in hismagnum opus, Popper conceded that which “basic statements” one chooses to accept as “true” or “false” was a matter of “decision”,  i.e. the problem of an ‘empirical basis’ isnot, to Popper, the same as that of an ‘observational basis’ (the two are commonly confused), and it is in respect to the latter that “decisions” in the conventionalist sense become important or even crucial. For this element of Conventionalism in Popper, see the index, s.v. “decisions”. Against Popper, Wartofsky,art. cit. argues that therecan be no “singular statements” of the sort required for Popperian refutations, as all language is “fraught with irremediable universality”, and withthis then compare Hegel, in the mesmerising sections of thePhenomenology that deal with the dialectic of sense-certitude.
  • 204.Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes”, Criticism and the Growth of Knowledge, eds. I. Lakatos and A. Musgrave (Cambridge 1970), p. 98. Much of the argument at this point is indebted to Lakatos’s attempt to salvage Popperianism against disintegration, Lakatos and Feyerabend being simply the two faces of neo-Popperianism.
  • 205.Lakatos, “Falsification”, pp. 106–7; cf. Popper, Op. cit. p. 50, “In point of fact, no conclusive disproof of a theory can ever be produced”.
  • 206.Schmidt, Op. cit. p. 118 more or less proposes this: “Marx agreed with the Enlightenment that thought which was not directed towards the accomplishment of practical tasks became merely whimsical”. His evidence for this is the Second Thesis!
  • 207.Piaget’s work, like Chomsky’s, has fundamental importance for Marxism. Lucien Goldmann was the first Marxist to appreciate its significance. As early as 1948–52, at a time when most Marxists would scarcely have shown any interest at all in such matters, Goldmann (Recherches dialectques, pp. 118-45) reviewed Piaget’s work in two short articles. The first of these, “La psychologie de Jean Piaget”, underlined what Goldmann himself saw as the main or possible points of convergence between the “dialectical materialism” of Marx and the genetic epistemology of Piaget. These Goldmann described as (1) Piaget’s own dialectical method in the analysis of intelligence, characterised by its double rejection of merely rationalist or merely empiricist accounts of human intelligence, and specifically by its rejection of “toutes les oppositions rigides – instinct-intelligence, pensée-action, norme-fait – sans cependant jamais tomber dans l’électisme” (p.122). (2) Piaget’s adaptivism, centred on the two fundamental notions in his work of “assimilation” and “accommodation”, and Marx’s conception of labour as a general human activity. “Le role de la ‘nature’, de la ‘matière’, de l’objet,est identique dans la psychologie de Piaget et dans le matérialisme historique” (pp. 122f.). (3) Piaget’s polemic against classical and Russellian logic in the direction of a “dialectical logic of totalities” (p. 125). And finally, (4) Piaget’s constructivism, or his conception that between “action”, “activity”, etc. and the general development of human intelligence ontogenetically, there is a close and inseparable link. Here, Goldmann himself would refer back to theTheses on Feuerbach and write that “Piaget arrives atabsolutely analogous conclusions” (p. 126). It is this last theme that Goldmann resumes in his second review, “L’epistémologie de Jean Piaget”, seeing in Piaget’s work a coalescence of Brunschvig’s contructivist view of “the mind” (and of Janet’s emphasis on the role of behaviour). In this review, Goldmann sees the major significance of the “dialectical tradition”, from Pascal through Kant and Hegel into Marx, and after Marx, into Piaget, in the two-sided polemic launched by it against both empiricism and rationalism, though he argues that “neither Hegel nor Marx ever succeeded in explaining in a theoretically satisfying way the double, deductive-cum-empirical, nature of thought” (p. 134).  Goldmann wrote these reviews at a time when Vygotsky’s work, banned by Stalin and hardly known in the West, was probably the only exemplar of a Marxist-genetic theory of concept-formation; and at a time when the major efforts towards a “Marxist psychology”, e.g. Pierre Naville,Psychologie, marxisme, matérialisme (1946), more or less consciously espoused Watson’s behaviourism, true to their own “materialism”. It was, of course, against the behaviourist accounts of language-acquisition that Chomsky’s own work was later directed, and it is to the early polemical responses to associationism and stimulus-response theories that Piaget’s constructivist views are finally traceable.
  • 208.A. Gramsci, “Il materialismo storico e la filosofia di Benedetto Croce”, Opere di Antonio Gramsci (Einaudi 1948), t. 2, p. 143.
  • 209.See Jacques Texier’s excellent introduction,   Gramsci (Paris, 1966), esp. pp. 68–78.
  • 210.J. Piaget, Main Trends in Psychology  (London, 1973) p. 45. Piaget makes it clear both in this work and elsewhere that he does not accept the hypothesis of “innateness”. InLe structuralisme (Paris 1968) p. 52 he even opposes the constructivist view of the growth of intelligence to innatist conceptions. This accounts for his cautiously, or implicitly, critical attitude to Chomsky (Op. cit. p. 63f.) and for Chomsky’s similar reaction to Piaget (Language and Mind, p. 80). However, this ambivalence is, to a large extent, one inherited from the traditional but now archaic opposition within psychology between preformationalism and interactionism (and that this is what Piaget still seems to have in mind is clear fromStructuralisme, p. 52, “Au total il n’y a que trois solutions – préformation, créations contingents ou construction”). Preformationism, on the other hand, arose and developed specifically as a deterministic, evolutionary, and biological theory of human intelligence, whereas the hypothesis of “innateness” was revived by Chomsky precisely to account for the creativity or spontaneity of language-acquisition. Secondly, the case of Lorenz himself should have suggested to Piaget that there is no necessary contradiction between positinga priori forms on the one hand, either biological, as with Lorenz, or mentally-specific, as with Chomsky, and retaining a basic role for “interaction”.
  • 211.J. Piaget, The Origins of Intelligence in Children (New York, 1952), where the subject of “curiosity” is dealt with in detail, through experiments on his own children. For the notion of “deductive constructions” which “experience” by itself cannot explain, cf.The Construction of Reality in the Child (London, 1954).
  • 212.Cited by J .Hunt, Intelligence and Experience (New York, 1961) p. 149 ff. However, Gestaltism, together with theDenkpsychologie of the Würzburg school, formed the first systematic critiques of sensationalist-empiricist psychology.
  • 213.Lukács, Geschichte und Klassenbewußtsein, p. 242;History and Class Consciousness, p. 132. When I first read this passage, I reacted quite sharply because I still had to distinguish the principle of experimentalism as a determination of human activity in general from scientific experiments in particular (of course, Lukács does not make any such distinction). In retrospect, with this distinction established, Lukács’s view becomes perfectly correct for scientific experiments as such.
  • 214.K. Marx, The Civil War in France (Chicago, 1934) p. 90.
  • 215.K. Marx, The Eighteenth Brumaire of Louis Bonaparte (Moscow, 1972), p. 14.
  • 216.See J. M. Cammett, Antonio Gramsci and the Origins of Italian Communism (Stanford 1967), p. 42, summarising the contents of “Socialism and Culture” (1916) (published in vol. 8 of the Einaudi collection).
  • 217.R. Luxemburg, “Organizational Questions of Social-Democracy”, in Rosa Luxemburg Speaks (New York 1970), p. 130.
  • 218.V.I. Lenin, “An appeal to the party by delegates to the Unity Congress who belonged to the former ‘Bolshevik’ Group”, Collected Works, vol. 10 (Moscow, 1965), p. 310; L. Trotsky,Terrorism and Communism (Michigan, 1961), p. 101.
  • 219.Capital, vol.1, p. 177 (Moscow edition), following the Moscow translation here against Fowkes, p. 283.
  • 220.Theories of Surplus Value, Part 2, p. 117.
  • 221.Grundrisse, p. 706, and cf. Marx’s expression “general intellect”.
  • 222.Grundrisse, p. 700.
  • 223.Grundrisse, p. 712, translating Ausübung as “practice” with Nicolaus rather than “exercise” with Fowkes (apud Schmidt).
  • 224.Early Writings, p. 251 (“Critique of Hegel’s Philosophy of Right, Introduction”) where Marx writes, “Theory is capable of gripping the masses when it demonstrates ad hominem”.
  • 225.Early Writings, p. 327.
  • 226.Capital, vol. 1 (Moscow tr.), p. 423.
  • 227.Ibid.
  • 228.Early Writings, p. 327.
  • 229.Early Writings, p. 327.
  • 230.T.S. Eliot, The Hollow Men; Baudelaire, “Teeming city, city full of dreams”; Eliot, “Unreal City / Under the brown fog of a winter dawn, / A crowd flowed over London Bridge, so many,/ I had not thought death had undone so many”; Benjamin on Baudelaire, “The Paris of his poems is a sunken city”; Shelley, “Hell is a city much like London—A populous and a smoky city”. Cf. also the wholemotif of “ghostliness” that dominates the writings of Kafka. In his conversations with Janouch, Kafka describes Taylorism as “defil[ing] and degrad[ing] not only the work but, above all, the human being who is a component of it”. “A Taylorized life is a terrible curse… The conveyor belt of life carries one somewhere, but one doesn’t know where. One is a thing, an object, rather than a living organism”  (Gustav Janouch,Conversations with Kafka, London 1953, p. 115) The contrast that Henri Lefebvre draws betweenpraxis andmimesis inMétaphilosophie. Prolégomènes (Paris, 1976) is brilliant. The reverse aspect of this Taylorised and automatic, active-passive, mimetic quality of life under urban capitalism is the displaced symbolism of creativity in literature and art, for example, Valéry’s poetry, where the sea comes to symbolise movement, unconscious and creative life, and even more obviously in the manipulation of language itself.
  • 231.Benjamin, “On Some Motifs in Baudelaire”, Illuminations, tr. Harry Zohn (New York, 1968), p. 175.
  • 232.In the Manifesto of Futurism which was first published inLe Figaro in 1909, it is impossible to grasp the deeper significance of futurism, surrealism and, more generally, of the renovated modernism of the twentieth century unless one explores the role that “automaticity” and “mechanism” play, both technically and symbolically, in these art-currents. If the fascination for gadgets, technical objects, machines, and plasticity in general was still an embryonic one in the bourgeois civilisation of Baudelaire’s day, then, by the First War, it became pronounced, exaggerated and almost obsessive.
  • 233.W. Benjamin “Über einige Motive bei Baudelaire”, Charles Baudelaire. Ein Lyriker im Zeitalter des Hochkapitalismus (Frankfurt 1974), pp. 127ff, tr. Charles Baudelaire (London 1973) pp. 132ff. where Benjamin explores the motif of the city and the crowd in great detail. The citations from Marx are from volume 1 ofCapital, the chapter on “Machinery and Modern Industry”. Benjamin understood the point made in the footnote above better than Lukács. That is why his literary appreciations are profoundly historical and lacking in any of the insipid, pietist moralism that permeates Lukács. Cf also Benjamin’s perceptive remarks on film in Das Kunstwerk im Zeitalter seiner technischen Reproduzierbarkeit (Frankfurt, 1963).
  • 234.Nicholas of Cusa, cited Colletti, Marxism and Hegel, p. 242.
  • 235.Capital, vol 3 (Moscow), p. 88; Grundrisse, p. 610f.
  • 236.Lukács, History and Class Consciousness, p. 89.
  • 237.Engels, Condition of the Working Class in England, Marx/Engels,Collected Works, vol 4 (Moscow, 1975), pp. 466, 415.
  • 238.Marx and Engels, German Ideology, p. 42.
  • 239.Early Writings, p. 256.
  • 240.Hegel, Science of Logic (Miller), pp. 783ff., 818f.;Logic (Wallace) pp. 261, 283f., 290f.
  • 241.From Kautsky’s article in Neue Zeit, 1901–2, cited Lenin, ‘What Is to Be Done?’,Selected Works, vol 1 (Moscow, 1961), p. 129;Collected Works, vol 5 (Moscow, 1961), p. 383.
  • 242.Lenin, “What Is to Be Done?”, Collected Works, vol.5, pp. 374ff. Of course, the Lenin who built the Bolshevik Party is not this Lenin, andcould not have been this Lenin. The conscious downgrading of class spontaneity is an element that rapidly fades from Lenin’s thought after 1905. This requires a separate investigation.
  • 243.This is brought out by Texier, Gramsci, who makes it the unifying basis of Gramsci’s philosophical and political conceptions. Very interesting is Texier’s comment on the way Gramsci understood theTheses on Feuerbach. “Gramsci regarded them as a basic text and referred to them constantly. However, he refused to interpret them in the Crocean manner to mean a supersession of philosophy in the name of praxis; rather he saw in them the affirmation of the unity of philosophy and praxis and consequently a new conception of philosophy itself” (p. 31). Again, “Does Marxism or the philosophy of praxis derive from the ‘realist’ conception [from materialism – JB] or is it, on the contrary, a continuation [prolongement] and critical surpassing [dépassement] of the “subjectivist” and idealist conceptions of being and cognition? To Gramsci the answer leaves no room for doubt. The philosophy of praxis – it is enough to recall the first twoTheses – derives from classical German philosophy, i.e. from idealism. Materialism fundamentally failed to grasp the nature of human activity … in failing to see reality as the result of man’s transformative action.It separated human knowledge from practical activity. Because it failed to grasp human creativity, both the nature of knowledge and that of truth remained impenetrable for it” (p. 65). In Togliatti’s writings on Gramsci, by contrast (cf. Togliatti,Gramsci, Rome 1967), this praxiological content loses all philosophical significance and degenerates into a politics of “national peculiarities”. For example, at p. 130, one reads, “Nel modo come Gramsci interpreta e rinnova la dottrina del marxismo rivoluzionario e quindi implicita l’affermazione della necessita della avanzata verso il socialismoper una via nazionale”.
  • 244.A. Gramsci, Prison Notebooks (London 1971), pp. 190, 334, 364f.
  • 245.This refers specifically to Luxemburg’s early essay, “Stagnation and Progress of Marxism” (1903). I have no idea if Luxemburg ever elaborated on the theme of “spontaneity” in more than a purely aphoristic and inchoate way.
  • 246.Lukács, “Moses Hess”, Op. cit., p. 195.
  • 247.Early Writings, p. 252.
  • 248.Early Writings, p. 201 (Marx to Ruge, May 1843).
  • 249.K. Marx, Critique of the Gotha Programme (Moscow 1971), p. 9.
  • 250.Marx, Op. cit., p. 29. This is already argued at length in the first volume ofCapital (Fowkes’s translation), pp. 613f. “As Robert Owen has shown us in detail, the germ of the education of the future is present in the factory system; this system, this education will,in the case of every child over a given age, combine productive labour with instruction and gymnastics, not only as one of the methods of adding to the efficiency of production, butas the only method of producing fully developed human beings”. And cf. “The possibility of varying labour must become a general law of social production, and the existing relations must be adapted to permit its realization in practice. That monstrosity, the disposable working population held in reserve, in misery, for the changing requirements of capitalist exploitation, must be replaced by the individual man who is absolutely available for the different kinds of labour required of him; the partially developed individual, who is merely the bearer of one specialised social function, must be replaced by the totally developed individual, for whom the different social functions are different modes of activity he takes up in turn” (p. 618). For the deeper significance of all this, cf.Theories of Surplus-Value, Part 2 , p. 118.