Book Review

The Humanism of Ernst Cassirer


If you find contemporary philosophy unappealing, this biography of Ernst Cassirer may bring you back to the discipline you once thought you might love when you read Plato’s Phaedo for the first time.[1] Many college students and much of the reading public are apparently put off by the cold detachment of Anglo-American logical positivism and its projects, which always begin with a working knowledge of predicate logic. The aim of Rudolf Carnap, Moritz Schlick, and Otto Neurath was to develop a purified formal language that would codify all (empirical) knowledge in a single scientific theory, and then to purge philosophy of metaphysics and metaphor. Predicate logic is a late-nineteenth-century algebraization of traditional logic, invented by Gottlob Frege and improved by Bertrand Russell and Alfred North Whitehead before and after 1900. It encompasses not only forms of Aristotle’s syllogistic, but also the propositional logic of the Stoics, and the nascent discipline of set theory. The logical positivists renovated the program of Comtean positivism by deploying the powerful and expressive new logic to “rationally reconstruct” everything from physics to sociology.

The other philosophical pole is equally unattractive. It stems from Lebensphilosophie in Germany and France, which also flourished before and after 1900 in the work of Friedrich Nietzsche, Henri Bergson, Oswald Spengler, and later Martin Heidegger. These thinkers mocked the pretension of reason (as well as the abstract rules of liberal democracy), which shuts us off from the abundance of life and reduces the powerful, cruel aristocrats of the human spirit to the paltry level of the bourgeois and quotidian. The influence of this philosophy ran deep in the twentieth century, because it attracted thinkers put off by logical positivism; but its irrationalism profoundly damaged the commitment of philosophy to liberalism and humanism. So, shall we reduce reason to logic, foreswear history and metaphysics, and reduce ethics to epistemology? Or shall we awaken Man to his own nothingness, and await the next disclosure of Being or the siren call of the transient moment? Neither option is my idea of a good time, and I’m a professor of philosophy. In sum, contemporary Western philosophy seems to have abandoned a politically and philosophically essential middle ground.

What shall we do? One constructive step would be to revive the study of works by philosophers who sought that middle ground but have now fallen out of fashion. A good candidate is Ernst Cassirer, whose Philosophy of Symbolic Forms (his three-volume masterwork, among his many other books) has been swept under dune grass by the hyper-rationalist and irrationalist philosophical tides. Cassirer opposed both Carnap and Heidegger in his brief professional ascendency, when he held a chair in philosophy at the University of Hamburg from 1919 until 1933. (See Michael Friedman’s thoughtful A Parting of the Ways: Carnap, Cassirer and Heidegger.[2]) In 1933, he emigrated from Germany because he was Jewish, spending the last eleven years of his life at Oxford University, the University of Göteborg (Sweden), and Yale University. He died in 1944, and his papers (available online) are housed in Yale’s Beinecke Library. Yale University Press publishes his most important books in English translation, though even in paperback they are very expensive. On Amazon.com’s “Greatest Hits” list, while the volumes of The Philosophy of Symbolic Forms stand between 500,000 and 1,000,000, Heidegger’s Being and Time stands at 30,000. The thirty-three contributors (a star-studded cast of analytic philosophers) to the volume in the Boston Studies in Philosophy of Science, In Memory of Rudolf Carnap (Springer, 2008), make the ten contributors (excellent scholars, but not household names in Anglo-American philosophy, since most of them are Israeli or French) to The Symbolic Construction of Reality: The Legacy of Ernst Cassirer look eccentric.[3]

When I was a graduate student in philosophy at YaleUniversity in the 1970s, I took plenty of courses in standard analytic philosophy and spent a whole semester on Heidegger’s Being and Time. Although I arrived at Yale with my lifelong topic “poetry and mathematics” already well formulated, nobody ever suggested that I use Cassirer’s work as a starting point for thinking about my double topic. Heidegger took poets seriously (with the notable exception of Paul Celan, as my colleague Veronique Foti points out in her monograph Heidegger and the Poets[4]), but he despised science. The logical positivists were deeply interested in science and mathematics (though they distorted them by subjecting them to formal logic), but what they had to say about poetry, and art generally, was silly. Here is Carnap, for example: “The aim of a lyric poem in which occur the words ‘sunshine’ and ‘clouds,’ is not to inform us of certain meteorological facts, but to express certain feelings of the poet and to excite similar feelings in us. A lyric poem has no assertional sense, no theoretical sense, it does not contain knowledge.” (Philosophy and Logical Syntax[5]). Tell that to William Empson, buddy. I truly don’t understand this collective amnesia, though Edward Skidelsky has an interesting explanation in his new biography, Ernst Cassirer: The Last Philosopher of Culture.

Both logical positivism and Lebensphilosophie were reactions against the neo-Kantian movement that arose in Marburg, heralded by Hermann Cohen’s first book in 1871—substantially revised in 1885—Kant’s Theorie der Erfahrung (Kant’s Theory of Experience). Kant defended Newton’s physics against the skepticism of Hume by grounding it in “pure understanding,” that is, he tried to show that the Newtonian axioms articulate certain conditions of possibility of any experience whatsoever, in his “transcendental deduction of the categories.” This position was hard to maintain in light of the revisions in physics and mathematics brought about by James Clerk Maxwell’s theory of electricity and magnetism and the new laws of thermodynamics, as well as the development of non-Euclidean geometry, which challenged Kant’s “transcendental aesthetic.” Beginning in 1905, the revolutions of special and general relativity, and then of quantum mechanics, intensified this difficulty. Neo-Kantianism, as Skidelsky points out, was faced with a choice: either become a theory of pure consciousness, or become a theory of science. Cohen took the latter route; Husserl took the former. And this decision made Cohen’s critical philosophy both historical and enclitic; it became the reflective study of the development of mathematics and science over time and lost its “foothold in eternity,” the priority that Kant claimed for the study of the transcendental that must always precede any investigation of the merely empirical. For Kant, nature was constituted by human understanding itself and thus amenable to codification by a complete set of principles (Newton’s mechanics), though to encompass the realm of morality and to tame metaphysics, the finitary understanding must be regulated by infinitary reason. For Cohen, by contrast, nature always surpasses human understanding, so that the work of science (and therefore philosophy) must be open-ended; the demand for universal validity remains the one fixed constraint on a progressive science, but it is merely regulative, not constitutive.

Proponents of the MarburgSchool, though devoted to science, regarded it as a “free and active creation of the intellect.” This set them at odds with earlier versions of positivism (Comtean, neo-Darwinist, Machian), which regarded science as the passive adaptation of the human mind to sensory data, a mechanism of survival. And they refused to sever the philosophical study of science from that of ethics, which the positivists rejected as subjective “poesy.” Both science and ethics they regarded as objective, expressions of the same spontaneity and creativity of human reason. The MarburgSchool was also at odds with Hegelian idealism, because it rejected the central tenet of Hegelian metaphysics, the identity of reality and thought. Ultimate reality always eludes human thought; philosophy must maintain its modesty and circumspection in the face of its infinite task. In sum, the MarburgSchool tried to humanize science, rationalize religion, and liberalize socialism. Nonetheless, its project was one sided, for Cohen held “the extreme view that experience is objective only insofar as it can be given mathematical form.” Science thus lost its foothold in everyday experience, and Cohen was never really successful in extending his critical method to nonscientific experience, aesthetic, ethical or political; in particular, he could not address the carnage of World War I, brought on in part by the advance of technology, the flower of scientific “progress.”

Ernst Cassirer was Cohen’s prize pupil. I recall the strange conventions of the German University system in 1976 when I watched fellow graduate students at the University of Muenster (Westphalia, Germany) meekly acquiesce to every stated opinion of their dissertation advisor or Doktorvater, not only in typescript but in private conversation; sixty years before the inhibition would have been all the more stringent. The only book of Cassirer’s I worked on seriously when I was in Muenster was Substance and Function, which never tempted me to pursue his other work; but it was written in 1910, under Cohen’s jurisdiction and moreover the spell of the new logic of Frege and Russell. Only in 1921, three years after Cohen’s death and when he was nearly fifty, did Cassirer first announce his own, independent “philosophy of symbolic forms.”

Substance and Function begins with a critique of Aristotle’s logic of the syllogism and doctrine of the abstraction of concepts. Syllogistic reflects a metaphysics of substance and attribute, for it only treats sentences of the form “S is P”; its formalism cannot express relations or functions. Aristotle understood knowledge as beginning in the perception of individual substances, like horses, people, or trees: the concept is a selection from what is immediately presented in experience. Every collection of comparable objects has a supreme genus consisting of all the properties in which those objects agree and eliminating all the properties in which they do not agree. As we go up the hierarchy of concepts, then, the content of the more and more generic concepts diminishes, and the category of Being seems to have no content at all. But this is a problem, because generic scientific and mathematical concepts are supposed to give us more and more precise determinations, that is, more content.

Cassirer’s central claim, based on his study of modern mathematics and physics, and the new logic, is that modern mathematical concepts do not cancel or forget the determinations of the special cases but fully retain them. When a mathematician makes his formula more general, Cassirer asserts, this means that he is able not only to retain all the more special cases, but also to deduce them from the universal formula. (If a general concept had been arrived at by Aristotelian abstraction, the special cases could not be recovered from it, because the particularities have been forgotten.) By contrast, the mathematical or scientific concept seeks to explain and clarify the whole content of the particulars by exhibiting their deeper systematic connections, revealed in the law of the series. Thus from Descartes’s general equation Ax2 + By2 + Cx + Dy + E = 0, we can elicit the particular conic sections (circle, ellipse, hyperbola), presented in systematic interrelation. Here the more universal concept is more, not less, rich in content; it is not a vague image or a schematic presentation but a principle of serial order.

The general is no longer seen as something over and above the “sum” of particulars, but immanent in them; it is the particulars viewed under the aspect of their serial form. Thus in modern mathematics, things and problems are not isolated, but shown to be in thoroughgoing interconnection. Moreover, Cassirer continues, the concrete universality of the mathematical function extends to the scientific treatment of nature. Thus a series of things with attributes is transformed into a systematic totality of variable terms or parameters; things are transformed into the solutions of complex equations, as when a molecule becomes the solution to a wave equation, or when the sun, the moon and the earth become a solution to the three-body problem. As Cassirer wrote earlier, the world of sensible presentations is not so much reproduced as supplanted by an order of another kind. But now we have returned to the central difficulty faced by the MarburgSchool: the realms of science and ordinary experience have been dissevered.

Cassirer was brought up on the poetry of Goethe and the anthropology of Herder, Schiller, Schelling, Hegel, and Humboldt. He shared the romantic opposition to eighteenth-century rationalism for degrading our “sensuous, emotional life to the level of a biological residue, a passive stuff to be overcome.” These thinkers defined the essence of humanity not as reason but rather our capacity for self-expression, manifest in not only science and mathematics, but also language, religion, art, and myth. The ways in which we articulate and organize the world are irreducibly plural. Cassirer learned from them that our relationship with the world is not dominated exclusively by the demand for “objective knowledge,” but must also answer to the human thirst for meaning, how we shape the world into patterns, our various activities of symbolic formation. “The critique of reason becomes, in Cassirer’s famous declaration, the critique of culture.” He did not, however, share the Romantics’ disdain for science. Rather, “Cassirer’s ultimate purpose was to reveal science as an expression of the same symbolic capacity underlying language, art, and myth, thereby acquitting it of the common charge of coldness and inhumanity. His philosophy is an attempt to exploit the ambiguous energies of German romanticism on behalf of enlightenment.”

Skidelsky argues that the presiding influence on Cassirer, the counterweight to Kant, Cohen and the logicians, was Goethe. From the great poet Cassirer learned to believe in the objectivity of the artistic imagination. Whereas for Kant, aesthetic judgment, the apprehension of beauty, is universally valid but still subjective, for Goethe beauty is a revelation of the real form of things; art as well as science is a mode of world constitution. Cassirer transforms his earlier account of the serial concept by detaching it from its mathematical context and presenting it as emblematic of the creative imagination. The poet, like the mathematician, “sees the individual not as an isolated, self-contained sub- stance but as the symbol of a more universal complex.” Thus Cassirer especially admired Goethe’s gift for disclosing the general in the particular.

In The Philosophy of Symbolic Forms (published between 1923 and 1929), Cassirer not only urges this similarity but also explores the radical differences among the world-makings of mythology, language, religion, art, and science. He gives up the Kantian (and Hegelian) devotion to a complete and unified system in favor of irreducible plurality: thus his masterwork has three volumes, each a fresh departure from the others in the exploration of how we human beings use symbolic forms to organize our world. This lack of strict unity means that the project of thinking them together—using irony, analogy, and deliberation—must continue indefinitely. An honest thinker must always reflect on the limitations of his or her own methods, “organs of interpretation,” and maintain flexibility. In Cassirer’s view, because critical philosophy defines objectivity not in relation to an object but rather in terms of the immanent lawfulness of thought or human awareness, more than one “objective” conception of the world can be admitted. Language, art, myth and religion, as well as science, are then different ways of accomplishing the synthesis of spirit and world. Both Goethe’s Farbenlehre and Newton’s arithmetization of color may be entertained: “The concept of the symbol is both broad enough to unite the various cultural forms and flexible enough to do justice to their individuality.”

In 1920, Cassirer came to know the Kulturwissenschaftliche Bibliothek Warburg, soon after he took up his professorship at the University of Hamburg, as well as Fritz Saxl, Erwin Panofsky, and later, in 1925, the Director Aby Warburg who had recovered from a long illness. Cassirer became the “house philosopher” of the Warburg Library; his influence is evident in two influential books: Saturn and Melancholy (1923) jointly authored by Saxl and Panofsky, and Panofsky’s Perspective as Symbolic Form (1927). The library itself was, as Skidelsky observes, the “objective correlative” of Cassirer’s own developing theory of symbolic forms, for its collection of books was presented as a system of symbols. “Iconography replaced authorship or chronology as the basic principle of organization. The aim was to make visible the fundamental symbolic patterns underlying the individual works of art and literature.” Warburg’s stormy personality, Skidelsky argues, was also a beneficial influence on the often too-serene, too-patrician Cassirer. Whereas Cassirer looked for mediation and conciliation among the symbolic forms, Warburg discerned the possibility of real conflict and so forced Cassirer to be more realistic in his assessment of cultural and political realities, a realism evident in his last book, The Myth of the State (1946), written in exile and published just after his death.

Both Cassirer and Warburg viewed myth as the unruly element in any system of culture, full of energy and ineluctable. Over-intellectualized systems (philosophical or political) that try to get rid of it entirely find that it returns with a destructive explosion; irrationalist doctrines that try to codify and use it for ulterior purposes find that although at first it lends energy to a program, in the end it proves anarchic and elusive. “Myth is both a source of creativity and a threat. It invigorates, but can also overwhelm.” Cassirer concludes that philosophy’s task is not to eliminate myth but to understand it, locating it within the whole (rich, plural, unstable) system of culture. He and Warburg were acutely aware of fascism as a re-eruption of myth, challenging the cool rationality of liberal democracy. (This assessment of myth is, I think, unfairly negative, and should be balanced by the positive view inherent in Gaston Bachelard’s later works, like The Poetics of Space.)

By symbolism, Cassirer meant a natural potency inherent in consciousness: “Thought does not flow here in a finished riverbed which has been made for it; rather, it must find its own way, it must dig its bed for itself.”[6] Our purchase on the world is always, from the beginning, active and objective; the way to arrive at a better understanding of ourselves is to study our ways of world-making. Thus philosophy has no empire of its own; its project is to reflect upon the arts and sciences, to be the self-consciousness of culture. This subjection of philosophy made Heidegger furious; and the “demotion” of mathematics and science was strenuously opposed by the logical positivists. Both Heidegger and Carnap sought a fixed location, outside of history, which philosophy could call its own. Thus too Cassirer’s work was an embarrassment to my professors at YaleUniversity; thirty years after his death, impelled by different fears, they all wanted to escape the objective plurality of symbolic forms.

Human beings confront the world already shaped, symbolically, by their own aspirations, projects and theories. For Cassirer, the symbol was an instrument of human liberation, and he therefore ranked the different symbolic forms according to their power to liberate. Myth occupied the lowest rung, because in myth “symbols are treated as objective powers; their source in human spontaneity is forgotten.” Having escaped enslavement to nature, he wrote, we re-enslave ourselves to custom. Religion was more highly regarded by Cassirer, for religion (especially the Judeo-Christian tradition and the ethical religion of Kant) allowed us to recognize and realize the autonomous, ethical will. So too the scientific disenchantment of nature was the other side of this process of human liberation. All the same, Cassirer’s view of the “system” was not a ladder, like the Great Chain of Being or Hegel’s progress of Spirit; it was more like a tree. Myth will never fully lend itself to liberation; religion typically relocates human autonomy in the otherness of God; science rejects the claims of myth altogether and belittles art, cutting itself off from life; art faced with science dissolves in irony.

What Cassirer warns against is one-sidedness, our tendency to elevate a single kind of symbolic form and deny legitimacy to others; his philosophy of symbolic forms seeks to explain this tendency to one-sidedness and also to offer strategies for overcoming it. The tyranny of science gives us logical positivism; the tyranny of myth gives us Lebensphilosophie; but the recognition of incommensurable symbolic forms demands of philosophy “a continually renewed effort of reflection.” Twentieth-century philosophy has been, on the whole, lazy. The equilibrium of civilization is a project always in need of renewed commitment, and its many unbalances, ever more striking as the twenty-first century begins, demand a fresh approach. The humanism of Ernst Cassirer, Erwin Panofsky, Gaston Bachelard, and Alexandre Koyré (for example) could help us regain the philosophical middle ground that we have for too long abandoned.


[1] ERNST CASSIRER: The Last Philosopher of Culture, by Edward Skidelsky. PrincetonUniversity Press. $28.63.

[2] Michael Friedman, A Parting of the Ways: Carnap, Cassirer and Heidegger (Chicago, 2000)

[3] Jeffrey Andrew Barash, The Symbolic Construction of Reality: The Legacy of Ernst Cassirer (Chicago, 2008).

[4] Veronique Foti, Heidegger and the Poets: Poiesis/Sophia/Techne (Amherst, N.Y., 1995).

[5] Rudolf Carnap, Philosophy and Logical Syntax (New York, 1937).

[6] The Philosophy of Symbolic Forms, ed. by D. P. Verene (New Haven, 1996),Vol. IV, pp. 4–5.