Heidegger’s Disavowal of Metaphysics

Taylor Carman
Barnard College


Until the mid-1930s, Heidegger used the word “metaphysics” with no pejorative implication; it simply meant philosophy. By about 1936, however, he began using the word to refer not to philosophy as a whole, but to a dominant tradition beginning with Plato and ending with Nietzsche. Metaphysics, he would now say, does not just happen to fail to address the question of being, but occludes it, concealing it and rendering it unaskable, virtually incomprehensible. Heidegger’s disavowal of the word “metaphysics” was in part a rhetorical response to Carnap, but it also marked the beginning of his substantive critique of “representational” or “calculative” thinking. Representational thinking aspires to comprehend entities as such and as a whole in their being. But the horizon or background against which such comprehension takes place cannot itself occupy a place in the totality of entities, so the metaphysical aspiration is forlorn. Heidegger’s later thought aims at an “overcoming of metaphysics” – not in Carnap’s sense, but rather to think not just the meaning of being, which is to say being understood as the being of entities, but the truth of being, that is, the way in which being as such manifests itself. Confusion about this change of direction in Heidegger’s later thinking has been generated in part by his own disingenuous attempts to rewrite the history of his own early philosophy in order to make it appear more consistent with his later critique of metaphysical thinking.

There has been extensive scholarly debate surrounding Heidegger’s self-described “turn” (Kehre) from the phenomenology of Being and Time (1927) to his later work, and broad disagreement about exactly what the turn was and when it occurred. On some accounts, it had already taken place by 1930, at which point Heidegger no longer believed that fundamental ontology opened the way to a general consideration of the meaning of being as such. According to Heidegger himself, by contrast, the turn was not a change in his own philosophical views at all, but an impersonal event of some larger significance in the history of thought. Accounts of the shift from the “early” to the “later” Heidegger have as a result never fully managed to disentangled two quite distinct issues: his abandonment of the project announced and commenced in Being and Time on the one hand, and the his critique of metaphysics on the other.

Whereas Heidegger says very little explicitly about his abandonment of the project of fundamental ontology, probably around 1930, his disavowal of metaphysics in the late ’30s is explicit and well documented. Prior to 1936, Heidegger used the word freely with no pejorative or even critical connotation. The first sentence of Kant and the Problem of Metaphysics (1929), for example, credits Kant with “placing the problem of metaphysics before us as a fundamental ontology,” which he in turn defines as “the metaphysics of human existence, required for metaphysics to be made possible.”1 Similarly, his famous 1929 inaugural lecture “What Is Metaphysics?” addresses that question not by examining and discussing metaphysics at arm’s length, as it were, but by “tak[ing] up a particular metaphysical question,” thereby “let[ting] ourselves be transposed directly into metaphysics.”2 Heidegger’s lectures of 1929–30 are entitled The Fundamental Concepts of Metaphysics, metaphysics being, he explains at the outset, “the central discipline in the whole of philosophy.”3 As late as 1935, in the lecture course later published as Introduction to Metaphysics, far from excluding the Presocratics at one end of the tradition and himself at the other, he says again, “Metaphysics is the name of the definitive center and core of all philosophy.”4

By 1936, however, he began using the word very differently, to refer not to the entire history of Western philosophy, but to a dominant tradition within that history, beginning with Plato and ending with Nietzsche. Metaphysics, he would now say, does not just happen to fail to arrive at the question of being, but systematically suppresses it, concealing it and rendering it unaskable, indeed virtually incomprehensible. Why the change?

Heidegger’s disavowal of the word “metaphysics” was at least in part a rhetorical response to Rudolf Carnap’s 1931 essay, “Elimination of Metaphysics Through Logical Analysis of Language,” which ridiculed the “question of being” (Seinsfrage) as a prime example of nonsense in violation of the rules of logical syntax. Heidegger insisted, on the contrary, that Vienna Circle positivists like Carnap were the real metaphysicians, for it was they who had reduced the question of being to mere gibberish by arbitrarily restricting meaning to the formal constraints of logic and mathematics on the one hand, and to the material constraints of empirical inquiry on the other. In redefining the term “metaphysics” in this way, Heidegger was also beginning to distance himself from Nietzsche and Nietzsche’s doctrines of will to power and eternal recurrence, in effect retreating from the charged blend of political and philosophical rhetoric with which he had been supporting the Nazi regime, not only during his year as rector of Freiburg University in 1933–34, but for several years thereafter (just how long is not entirely clear).

What, then, did Heidegger think metaphysics was? A first, crude approximation is to say that metaphysics was, for him, knowledge of entities (das Seiende) as a whole, as opposed to being (Sein). But that is not quite right, for two reasons. First, for Heidegger until 1936, metaphysics, precisely by being knowledge of entities as a whole, is thereby – implicitly or explicitly, directly or indirectly – knowledge of entities as such, as entities, which is to say, in their being. An understanding of the totality of entities presupposes an understanding of being. Absent an understanding of being, there can be no understanding of entities, let alone entities as such and as a whole. In raising the question of being explicitly, then, Heidegger did not take himself to be introducing an altogether new question into a tradition lacking it entirely. Instead, he saw himself as uncovering a question that lay dormant in Western philosophy, reminding it, as it were, of the question that had defined it all along, since its inception, but that it had forgotten, at least since Plato. As we have seen, as late as 1935 Heidegger was presenting the question of being as belonging to and as motivated by the tradition, indeed as the culmination and fulfillment of that tradition, so he was perfectly happy to call his own thinking, and likewise that of the Presocratics, “metaphysics.”

The second reason it’s not correct to say simply that metaphysics is, for Heidegger, the knowledge of entities as a whole to the exclusion of being is that after 1936 Heidegger himself began to doubt, as he had not done previously, the intelligibility of the very idea of a knowledge of entities as a whole. Or so I shall argue. By the time of the Beiträge (1936–38), that is, metaphysics is, for Heidegger, not so much knowledge of the totality of entities as the deeply misconceived and forlorn aspiration to such knowledge. There is, he came to believe, something incoherent in the very notion of a knowledge of entities as a whole, so that metaphysics according to his earlier conception of it must be strictly speaking impossible. Of course, metaphysics, now understood as the misbegotten effort to know the totality of entities as such, is still possible – just as it’s possible to try to pull yourself up by your bootstraps. According this new conception, metaphysics is not a kind of knowledge at all, but a style of thinking, a way of understanding the totality of entities as conforming or corresponding to a kind of cognition or attitude – for Plato, intuitive apprehension of forms; for Descartes, rational certainty; for Nietzsche, perspectival will to power – that grasps those entities fully, adequately, as such and as a whole.

Specifying the precise difference between Heidegger’s views before and after 1936, however, is no simple task. For one thing, it is not obvious that in the 1920s and early ’30s he accepted as possible what I believe he came to reject as incoherent in the late ’30s and thereafter. What he certainly accepted before 1936, and perhaps afterwards too, was the very idea of a totality of entities as such. Indeed, the one thing we can evidently mostly safely say about the sum total of occurrent entities is precisely that it is.5 Further, Heidegger seems to have supposed that human understanding can grasp such a totality – not, to be sure, by possessing complete knowledge of it, but simply by apprehending it in a primitive kind of thought. At a minimum, that is, we have an understanding – and moreover an affective apprehension – of entities as such and as a whole simply by grasping the (admittedly vague) concept everything. Moreover, not only is the very idea of a totality of entities intelligible, but Heidegger also seems to have believed that there is such a totality, at least understood as a sum total of “occurrent” (vorhanden) objects, as opposed to human beings and “available” (zuhanden) cultural artifacts, whose being is constituted by our understanding of them. That is, Heidegger held not only that there is a totality of occurrent entities, but that, unlike Kantian things in themselves, those entities have a determinate causal structure in space and time, a structure that is the way it is independently of us and our ways of making sense of it.

That last claim is what I have elsewhere called Heidegger’s ontic realism.6 Ontic realism is more robust than Kant’s “empirical realism,” which he offers as a corollary to his transcendental idealism, but it is not as ambitious as other forms of metaphysical realism, for it concerns only the ontological status of occurrent entities, not any actual or possible description or theory of them. Heidegger never supposed, that is, that there could be, even in principle – even, as it were, in the mind of God – a complete knowledge of everything. Unlike Kantian things in themselves, the reality about which Heidegger was an ontic realist is not an object correlative to omniscience, a notion Heidegger (rightly) rejected as incoherent. Put slightly more technically, I think Heidegger never believed in the existence of what are sometimes called facts (or propositions), such that the totality of entities must include a subtotality of facts (or propositions) that might then be the object (or content) of a complete knowledge of everything. So, even if Heidegger later became even more hostile to the notion of a complete knowledge or description of the totality of entities (even merely qua occurrent), that by itself does not constitute a sharp break from his earlier view.

Neither is it obvious that Heidegger ever gave up the idea that there is a totality of entities. If the metaphysical pursuit of knowledge of that totality is incoherent, it is not because the very idea of a such a totality is incoherent. What Heidegger came to regard as incoherent, I believe, was rather the idea of a knowledge of entities as such as a whole, that is, as entities. For such a knowledge would have to include a knowledge of knowledge itself standing in relation to that totality, as understanding it as such, and consequently as involving – indeed, resting on – a prior, more fundamental understanding of being. A genuine knowledge of entities as such as a whole, that is, must necessarily include in itself a further understanding of being. That was precisely what Heidegger claimed for his own philosophical project in Being and Time and immediately thereafter – namely, a continuation and radicalization of metaphysics, proceeding from what he called “traditional ontology” toward his own fundamental ontology, which would spell out the conditions of the intelligibility of our understanding of entities as a whole, culminating in a fully general account of the meaning of being. Like the logical positivists, Heidegger came to believe that modern science had superseded and absorbed, precisely by attaining an objective knowledge of nature in its pure occurrentness – but, crucially, without also grasping the being of occurrent nature as such. Thus in his 1964 lecture “The End of Philosophy and the Task of Thinking,” Heidegger notes the de facto “development (Ausfaltung) of philosophy into the independent sciences,” and even the “dissolution (Auflösung) of philosophy into the technologized sciences.” What the sciences cannot do is grasp being – hence Heidegger’s famously provocative, if potentially misleading, quip that “The sciences don’t think.”7

The change in Heidegger’s conception of metaphysics was both terminological and substantive. Terminologically, the word “metaphysics” took on a different meaning in his vocabulary before and after (roughly) 1936. Before then, it appeared in the titles of many of his lectures and books, and it just meant philosophy. As he says in Introduction to Metaphysics of 1935, “Metaphysics is the name of the definitive center and core of all philosophy” (EM 13). Afterwards, it referred just to that segment of the tradition that began with Plato and ended with Nietzsche, excluding the Presocratics at the beginning and Heidegger himself at the end.

Substantively, Heidegger came to see his own project as more radically discontinuous with the philosophical tradition. Specifically, he drew a sharper distinction between his own thinking and metaphysics in the second, narrower sense of the word. Whereas earlier he had understood Western philosophy as a whole, including the Presocratics, as having failed to make the question of being explicit and thematic, he came to regard the peculiar style of thinking that began with Plato and ended with Nietzsche as systematically incapable of even acknowledging, let alone addressing, the question.

One telltale text is the Introduction to Metaphysics – or rather, the published edition of 1953, which in addition to the original 1935 lecture course includes several supposedly clarificatory insertions.8 The longest, near the beginning of the book, is a (rather muddled) excursus explaining the original lecture’s characterization of metaphysics, which Heidegger now says was sketchy and misleading – but, he insists, deliberately so!

The lecture course begins with what Heidegger calls “the first of all questions” (EM 1). The question “first in rank for us as the broadest, the deepest, and finally the most originary question” (EM 2) is, Why is there something rather than nothing? This is not the same as the even deeper question concerning the meaning of being, but it presupposes it and, if we follow Heidegger, leads back to it. A few pages later he therefore says, “So, it turns out, the question, Why are there entities at all instead of nothing? forces (zwingt) us to the prior question, What about being (Wie steht es um das Sein)?” (EM 25). Notice that that assertion – that the question concerning entities as a whole forces us on to the question of being – flatly contradicts Heidegger’s later critique of metaphysics, according to which (as we have seen) “the light of being … no longer comes within the range of metaphysical thinking” (Wegmarken, 195).

For its part, Heidegger says in 1935, “The question we have identified as first in rank – Why are there entities at all instead of nothing? – is the fundamental question of metaphysics” (EM 13). The question of being as such, however, has never been asked explicitly: “In the treatise Being and Time the question concerning the meaning of being is posed and developed specifically as a question for the first time in the history of philosophy” (EM 64). Still, what Heidegger is doing in the 1935 lectures is introducing his students to philosophy as he himself understands it and practices it. As he says, “Metaphysics is the name of the definitive center and core of all philosophy” (EM 13).

This is the point at which Heidegger inserts a rather convoluted addendum in 1953. “For this introduction,” he writes, “we have intentionally presented all this in a cursory and thus basically ambiguous way” (EM 13, emphasis added). Heidegger is certainly right that ambiguities have crept into the lecture. A few pages earlier, for example, he had said, “Φύσις is being (Sein) itself, by virtue of which entities first become and remain observable” (EM 11). On the very next page, however, he says, “Entities (das Seiende) as such and as a whole the Greeks call φύσις” (EM 12). So, is φύσις being or entities? Considering that the ontological difference between being and entities is virtually the cornerstone of Heidegger’s entire philosophy, this is an astonishing inconsistency. And it raises further questions. For example, when the Presocratics said φύσις, what did Heidegger think were they thinking? Being or merely entities? And were they thinking “metaphysically”? Or, as Heidegger would later maintain, did metaphysics begin only with Plato? This degree of equivocation, it seems to me, is unaccountable absent a fundamental shift in Heidegger’s understanding of metaphysics, including (but limited to) the meaning of the word “metaphysics,” not long after the 1935 lectures. By the time he writes the 1953 addendum, at any rate, Heidegger is acutely aware of the ambiguity and attempts to explain it away:

According to our elucidation of φύσις, it means the being of entities. If the question is περὶ φύσεως, about the being of entities, then the discussion of φύσις, “physics” in the ancient sense, is already beyond τὰ φυσικά, beyond entities and is concerned with being. “Physics” determines the essence and the history of metaphysics from the inception onward. (EM 14)

In Aquinas, in Hegel, in Nietzsche, he continues, “metaphysics steadfastly remains ‘physics.’ The question concerning being as such, however, is of a different essence and a different provenance” (EM 14).

This is no minor refinement. Here, in 1953, in stark contrast to the 1935 lecture, Heidegger draws a categorical distinction between the question of being that was alive in, indeed at the very heart of, the metaphysical tradition (the question of the being of entities), and a different question of being that was not (the question of being as such). In 1953, that is, Heidegger presents himself retrospectively as asking a question that falls outside of metaphysics altogether. Plato, Aristotle, Aquinas, Hegel, and Nietzsche had all been doing “physics” in the ancient sense, that is, thinking the totality of entities – albeit, entities in their being. Heidegger is now not just carrying that tradition forward by making the question already inherent in it clearer and more explicit; he is asking a different question altogether, a question metaphysical thinking does not and indeed cannot ask.

“To be sure,” he continues, “within the purview of metaphysics, and if one continues to think in its manner, one can regard the question concerning being as such merely as a mechanical repetition of the question concerning entities as such” (EM 14). But this is a mistake, in which “the question of being as such is misconstrued as coinciding with the question concerning entities as such” (EM 14).

Heidegger then adds, “The ‘introduction to metaphysics’ attempted here keeps in view this confused state of the ‘question of being’” (EM 14). What is the “confused state of the question,” which Heidegger says the lecture course “keeps in view”? Considering its fundamental importance to him, it seems inconceivable that Heidegger himself would lose sight of the ontological difference between being and entities altogether. The confusion must instead have to do with what precise relation obtains between being and entities. Metaphysics, he says, in 1929 and in 1935, is concerned with entities as such and as a whole, or entities in their being, or even, as he sometimes allows, the being of entities. That is, it thinks the essential relation between entities and their being. Posing the fundamental question of metaphysics, the question concerning entities as such and as a whole opens up, leads to – indeed “forces” (zwingt) us on to – the question that is, as it were, just waiting to be asked, namely, What about being? Or, as he puts it in Being and Time, what is the meaning of being?

Within a few years of the 1935 lectures, however, Heidegger insists that metaphysics does not – indeed cannot – pose the question of being. Metaphysics is not a path or a bridge from the question concerning entities to the question of being, but an obstacle, a blind spot, an eclipse of the question. The way he puts this in 1953 is to say that although, as he had said before, metaphysics thinks entities in their being, or even the being of entities, what it does not and cannot think is “being as such” (EM 15).

In the 1953 addendum Heidegger makes it sound as if the 1935 lectures were merely reflecting a confusion inherent in the tradition. What has in fact happened is that he himself has in the meantime abandoned the concept of metaphysics on which the lectures were based, distinguished his own question more sharply from that of the tradition, and relegated metaphysics to a narrower domain in which it thinks entities (or perhaps “entities in their being”), but cannot think being (or perhaps “being as such”). In 1935 Heidegger very clearly credited the entire philosophical tradition – not just as far back as Plato, but including the Presocratics – with asking about entities as such and as a whole, which is to say, entities in their being. This is why in 1935 he was able to slide so easily back and forth between saying on one page that φύσις means “being itself” (EM 11) and on the very next page that it means “entities as such and as a whole” (EM 12).

Within just a few years of the 1935 lectures, that ambiguity had become intolerable, since it left no room for Heidegger to distinguish himself so categorically from the tradition, as he now very much wanted to do. Rather than owning up to the change, however, Heidegger instead maintained that his own question of being, even in Being and Time, had been nonmetaphysical from the outset.

1 Kant and the Problem of Metaphysics, 5th ed., R. Taft, trans. (Bloomington: Indiana University Press, 1997), 1.

2 “What Is Metaphysics?” D. F. Krell, trans. Pathmarks, W. McNeill, ed. (Cambridge: Cambridge University Press, 1998), 82.

3 The Fundamental Concepts of Metaphysics: World, Finitude, Solitude, W. McNeill and N. Walker, trans. (Bloomington: Indiana University Press, 1995), 1.

4 Introduction to Metaphysics, 2nd ed., G. Fried and R. Polt, trans. (New Haven: Yale University Press, 2014), 19 [13] (translation modified).

5 Consider, by contrast, Markus Gabriel, who denies that there can be any such thing as a totality of entities. See his Fields of Sense: A New Realist Ontology (Edinburgh: Edinburgh University Press, 2015) and Why the World Does Not Exist, G. S. Moss, trans. (Cambridge: Polity, 2015). For a critical review of Fields of Sense, see my “Gabriel’s Metaphysics of Sense,” The Harvard Review of Philosophy, vol. 23 (2016): 53–9.

6 See my Heidegger’s Analytic: Interpretation, Discourse, and Authenticity in “Being and Time” (Cambridge: Cambridge University Press, 2003), chapter 4.

7 What Is Called Thinking? J. Glenn Gray, trans. (New York: Harper & Row, 1968), 8. What Heidegger meant by this emerges more clearly much later in the text when he says, “Science does not think in the sense in which thinkers think” (134).

8 The most famous among them is his attempt to explain away the obviously jingoistic reference to “the inner truth and greatness” of National Socialism as a supposedly dispassionate comment on the growth of modern technology (EM 152).

beyng.com: the hyper-links to EM (Einführung in die Metaphysik) citations are linked to the corresponding page in GA 40.