Chapter 1 - The Two-Strand Method of Interpreting Kant

Summary

Chapter 1 argues that Heidegger, like many reconstructive interpreters, takes up the main question posed by the first Critique and attempts to identify Kant’s most plausible line of response to it, consulting the claims in Kant’s text alongside Heidegger’s own beliefs. Because Heidegger seeks to attribute true claims to Kant, his method of interpretation resembles that of Davidson and Gadamer. However, Heidegger improves on their method, because he recognizes the methodological role of disagreement in coming to agree with some author, thereby making room for differences in view between interpreter and text. He argues that we should expect great thinkers to struggle with their subject matter, offering competing strands of argument as they attempt to work out their view. The interpreter, therefore, must isolate the most promising strand of argument, differentiating it from less compelling arguments. Accordingly, Heidegger offers a two-strand interpretation of Kant that differentiates an insightful line of argumentation prioritizing imagination from a less promising line prioritizing understanding. Further, Heidegger offers a theory of error explaining why Kant struggles with his subject matter: Kant retreats to his less compelling argument due to the anxiety he experiences in uncovering the fundamental structure of the human being.

Many historians of philosophy interpret texts reconstructively. That is, many historians of philosophy attempt to reconstruct the best or most compelling account of the subject matter that the author puts forth, rather than taking on every claim that a text has to offer; they leave aside the claims that are unconvincing or even in tension with the most promising account. Reconstructive interpreters, then, are guided by concerns that are both exegetical (what is said in the text?) and philosophical (what is true about the author’s subject matter?). This method of interpretation is rightly called a charitable one because it takes seriously the author’s attempt to discover the truth about some subject matter; reconstructive interpretation attempts to piece together that true account. Further, reconstructive interpretation opens up the possibility of learning from the text – that is, learning the truth about the subject matter under discussion, by reference to the author’s most compelling account of it.

Despite the benefits of reconstructive interpretation, this method of interpretation comes with a risk. As Robert Brandom notes, interpreting a text in light of one’s own sense of what is true runs the risk of “hermeneutic ventriloquism” – “when the author’s lips move, but only the reader’s voice can be heard” (2002: 90). Interpreting a text in reference to one’s own beliefs about what is plausible might result in a reading where the interpreter fails to recognize meaningful differences between the views of the author and that of the interpreter; instead, the interpreter reads her own views into the words of the author. Hermeneutic ventriloquism undermines both of the benefits mentioned above: one is not being charitable to the author, as the author’s views have fallen quite out of the picture, and one does not learn from the text, but articulates one’s own beliefs.

Heidegger’s interpretation of Kant’s Critique of Pure Reason might seem like a prime example of hermeneutic ventriloquism. Indeed, the standard reception of Heidegger’s interpretation sees him as reading his own views into Kant. Ernst Cassirer may have been the first to offer this line of criticism, suggesting in a 1931 review that Heidegger’s Kant and the Problem of Metaphysics “penetrates … by force of arms into the Kantian system in order to subdue it and make it serviceable for his problem” (1967: 149). Similarly, Dieter Henrich argues that the interpretation is motivated by Heidegger’s own research program from Being and Time, having “little in common with the problems that determine the development of Kant’s thinking” and “results [that] diverge from the explicit position of the Kantian text” (1994: 53).¹ Scholars of Heidegger have largely conceded this point. For example, Daniel Dahlstrom comments that Heidegger’s “so-called ‘thoughtful dialogues’ often seem much more like rapacious monologues” (1991: 331).² Similarly, most book-length treatments of Heidegger’s Kant interpretation examine the interpretation in order to shed light on Heidegger’s thought, rather than taking it seriously as a way of interpreting Kant.³ There is widespread scholarly agreement, then, that Heidegger’s interpretation of Kant is more about Heidegger than it is about Kant; Heidegger engages not with Kant’s problems but his own, resulting in a reading of Kant where Kant’s doctrine looks remarkably similar to that of Heidegger.⁴ Indeed, Heidegger’s own admission that his interpretation of Kant is a “violent” one seems to vindicate this standard reading (PIK 247; KPM xx).

Heidegger certainly takes a reconstructive approach to interpretation – interpreting Kant in light of what Heidegger takes to be a plausible response to Kant’s leading question – but, I will argue, and against the consensus view, that his approach differs from reconstructive approaches that do run the risk of ventriloquism. While reconstructive interpreters like Donald Davidson and Hans-Georg Gadamer both seek widespread agreement with the authors who they interpret, Heidegger makes no such attempt. In fact, Heidegger claims to find a deep divide in Kant’s text: two strands of argument in tension with one another. There is a strand of argument with which Heidegger agrees, which prioritizes Kant’s faculty of imagination, and a less plausible strand of argument that prioritizes the faculty of understanding. Closely investigating the tensions in Kant’s text, Heidegger avoids an interpretation of Kant where Kant simply says what Heidegger would like him to say.

Contrary to the standard reception of the interpretation, Heidegger’s interpretation of Kant provides a model for navigating the dangers of reconstructive interpretation – that is, for avoiding the risk of hermeneutic ventriloquism. Interpreting Kant, Heidegger provides a model of charitable, reconstructive interpretation that nonetheless recognizes differences of view between interpreter and text. I outline Heidegger’s interpretative method in Section 1 of the chapter, showing that Heidegger attributes error as he interprets. In Section 2, I illustrate Heidegger’s interpretive method with an example: his interpretation of Kant’s Metaphysical Deduction. In Section 3, I argue that Heidegger supports his attributions of error with a theory of error, based on his account of anxiety in Being and Time. Finally, in Section 4, I address Heidegger’s later comments criticizing the violence of his late 1920s Kant interpretation, arguing that they do not provide compelling reasons to reject the interpretive method that he relies on therein. Together, these sections correct the standard reading of Heidegger’s method of interpreting Kant and bring out Heidegger’s contribution to the method of reconstructive interpretation.

1 Heidegger’s Reconstructive Interpretation of Kant

Heidegger often contrasts his own, reconstructive interpretive style – what he calls a “thoughtful dialogue” (denkendes Zwiegespräch) – with a more scholarly interpretive style that he dubs “historical philology” (historische Philologie) (KPM xx; see also KPM xvii). This distinction is comparable, I argue, to Brandom’s distinction between interpreting de re and de dicto (2002: 94–111). That is, the distinction hangs on whether the interpreter’s own beliefs about the subject matter under consideration in the text inform her interpretation of it. In the philological or scholarly method of interpretation that Heidegger eschews (Brandom’s de dicto interpretation), the interpreter’s beliefs about the subject matter are to play no role in interpreting the author’s claims. Rather, the interpreter attempts to interpret claims in light of other elements in the text (or the author’s broader corpus). This is not a dialogue but a rather one-sided affair; the interpreter merely attempts to get the author’s account right, bracketing her own beliefs about whether the account is a plausible one.⁵ As Heidegger remarks in a note on the Kant book, “discovering ‘Kant in himself’ is to be left to Kant philology” (KPM 175).

By contrast, in reconstructive interpretation (Brandom’s de re interpretation), the interpreter interprets in light of her own beliefs about the subject matter, consulting the claims that appear in the text in addition to her beliefs. This sort of interpretation is a dialogue: a conversation between author and interpreter about the subject matter treated in the text. Further, the dialogue is driven by a shared question. The interpreter enters into the dialogue because she is occupied with the same issues as the author; reading the text, she attempts to identify a plausible response to their shared question.⁶ Therefore, the interpreter consults her own beliefs in order to ascribe true claims to the author. This is the interpretive style, I argue, that Heidegger takes up in his interpretation of Kant.

More schematically, this method of reconstructive interpretation can be broken down into several components: First, when one interprets in this manner, one takes up the question posed by the author, and pursues it with the author over the course of the interpretation (Brandom 2002: 107). Second, one takes up the theoretical machinery offered by the author to answer this question – making use of the various concepts that the author introduces over the course of her inquiry. Third, the interpreter seeks to attribute to the author a plausible answer to the question that she poses, based on the interpreter’s own sense of what is plausible (Brandom 2002: 110). Gadamer is a proponent of this method of interpretation, suggesting that when we interpret a text, “we try to understand how what [the author] is saying could be right” (Gadamer 2013 [hereafter TM]: 254).⁷ Likewise, Davidson offers a principle of charity instructing the interpreter to read her own “standards of truth” into the claims that she is interpreting (2001a: 148).

As Cristina Lafont points out, these leading proponents of reconstructive interpretation share a fourth interpretive commitment: they suggest the interpreter ought to maximize truth, guided by the interpreter’s own beliefs concerning what is true about the subject matter (2008: 21). As Davidson puts this point, an interpretation ought to maximize agreement between the interpreter and speaker: “we want a theory … that maximizes agreement, in terms of making [a speaker] right, as far as we can tell, as often as possible” (2001b: 136).⁸ This method of interpretation tracks the author’s claims, and what can plausibly be claimed about the subject matter (according to the interpreter), so the interpreter can attribute the maximum number of plausible claims to the author. Similarly, Gadamer suggests that we ought not only to “understand how what [the author] is saying could be right,” but also to take the text to have a “unified meaning” (TM 256) or “assume its completeness” (TM 255). The task of interpretation is to determine how the text, as a whole, can be right about its subject matter. Indeed, Gadamer suggests that attributing an error to an author represents a failure of interpretation – a failure to determine how the text as a whole is right about its subject matter (TM 256). Gadamer, in his own terms, repeats the imperative to maximize truth when interpreting.

Lafont identifies Gadamer’s commitment to maximizing truth as a “methodological disadvantage” of his approach to interpretation, making it difficult to attribute and explain error (Lafont 2008: 22).⁹ The commitment to maximizing truth disinclines the interpreter from recognizing the author’s mistakes – from recognizing when the author holds different views from the interpreter. A difference in view is a bad result, to be avoided whenever possible. But authors can and do make mistakes; an interpretive method should be able to identify such mistakes. Further, without a way to mark disagreement between interpreter and author, reconstructive interpreting runs the risk of hermeneutic ventriloquism. I will argue that Heidegger abandons this fourth interpretive commitment; in his interpretation of Kant, he does not attempt to maximize the truth. I suggest that, in doing so, Heidegger’s reconstructive interpretation of Kant avoids the risk of hermeneutic ventriloquism.

We can clarify more precisely Heidegger’s departure from reconstructive interpreters like Gadamer and Davidson by shifting to the language of agreement. All three interpreters interpret with the aim of establishing agreement – that is, arriving at the author’s most compelling take on the subject matter, as judged by the interpreter. However, they depart in their understandings of how to achieve that aim. Gadamer endorses reading the text as consistently or coherently providing a compelling account; thus, many localized agreements between author and interpreter achieve large-scale agreement, about the account as a whole. By contrast, Heidegger does not demand consistency from the texts that he interprets. He expects serious philosophers – especially those posing the ontological questions that he and Kant pose – to struggle with their difficult subject matter, making some claims that are promising and others that are less so. Thus, he prescribes the same struggle to interpreters, who must isolate an author’s most promising account, differentiating it from those claims that are not quite there yet. Heidegger therefore recognizes the methodological role of disagreement in establishing agreement between interpreter and author. In his view, disagreement aids the interpreter in arriving at the author’s most compelling account of her subject matter. Thus, disagreement is not a shortcoming or a failure (pace Gadamer) of an interpretive method that attempts to establish agreement. Reaching an author’s deepest and best answer to her question calls on the interpreter to disagree with those claims that get in the way of her best account.

I will return to Heidegger’s departure from Gadamer and Davidson shortly. But first, I would like to introduce some textual evidence showing that Heidegger indeed employs the interpretive method that I have so far attributed to him.

Throughout his interpretation, Heidegger orients himself toward Kant’s main question and seeks to locate the most plausible answer to that question. In particular, Heidegger takes up “the real problem of pure reason” stated at the outset of the Critique: “How are synthetic judgments a priori possible?” (B19). That is, how can we draw conclusions about the objects we experience independently of our encounters with particular objects? Heidegger claims that “this possibility of ontological understanding, the possibility of a nature in general, the possibility of synthetic knowledge a priori, is the guiding ontological problem overall of the Critique” (PIK 276), and commits himself to pursuing this “actual” or “genuine problem” even when Kant seems to stray from it (PIK 209).

Given that Kant attempts to answer his question by inquiring into the various capacities of the human knower and their interrelationships, Heidegger takes it that answering Kant’s question requires inquiry into the “constitution” of the human being (PIK 224).¹⁰ After all, this constitution enables the human being to make synthetic judgments a priori. Heidegger is interested in this inquiry, as well. Indeed, he uses the term “fundamental ontology” to describe this type of investigation, both his own task in Being and Time, as well as that of Kant in the first Critique (BT 13; KPM 1). Both look to the constitution of the human being as a foundation for other ontologies, such as the ontology of natural objects. Pursuing this inquiry with Kant, Heidegger adopts Kant’s theoretical machinery; somewhat jarringly for a seasoned reader of Heidegger, Heidegger embraces discussion of the faculties as he pursues Kant’s answer to this question.

I have argued that Heidegger takes up Kant’s inquiry, interpreting Kant in hopes of answering their shared question. But two objections may occur to the reader at this point. The first objection starts from Heidegger’s own stated intentions in the Kant book. In his preface to the first edition, Heidegger says that the book offers a “‘historical’ [geschichtliche] introduction of sorts to clarify the problematic treated in the first half of Being and Time” (KPM xix). Is this not a straightforward admission that Heidegger does not deal with Kant’s “problematic” but his own? While Heidegger’s interpretation of Kant is indeed motivated by his own philosophical project, this does not mean that he got Kant wrong by reading himself into Kant. Heidegger selected Kant as the subject of his historical introduction to Being and Time for a reason, and the evidence that I have introduced above reveals that reason: Kant is the subject of Heidegger’s historical introduction, because Kant is engaged in the same endeavor – a fundamental ontology. Kant offers a good way to introduce Being and Time, since Kant also attempts a fundamental ontology.

A second objection starts from the term “anthropology.” While Heidegger repeatedly denies that his own philosophical project is an anthropology,¹¹ Kant occasionally adopts this label. As Heidegger recounts, Kant claims in his Lectures on Logic that answering the question “What can I know?” (i.e., the question posed in the first Critique) is anthropology insofar as it is related to the further question “What is man?” (9:25; PIK 48; KPM 146). If Kant accepts the label of anthropology and Heidegger rejects it, shouldn’t Heidegger differentiate their inquiries?

In fact, Heidegger brings up Kant’s claim precisely to disagree with Kant’s use of this label; Heidegger is adamant that Kant (like Heidegger himself) does not carry out an anthropology. In his 1927–1928 lecture course on Kant, Heidegger suggests that anthropology refers to an “empirical-ontic” inquiry (PIK 48); in the 1929 Kant book, he revises this definition to suggest that anthropology refers to an inquiry that is indeterminate, both in terms of what information it seeks about the human being and in terms of how that information bears on philosophical questions (KPM 147–148). Yet Heidegger takes issue with both senses of the term: anthropology in the former sense does not accurately describe Kant’s inquiry, and anthropology in the latter sense is unhelpfully ambiguous. Heidegger suggests that Kant does not seek an empirical account of the human being, as such an account would not explain how we make judgments independently of experience (PIK 50). Rather, Kant seeks (like Heidegger himself) “the ontological and essential structure of Dasein, the transcendental constitution of the subject” (PIK 224) – the constitution that enables the subject to make synthetic judgments a priori. For this reason, Kant’s inquiry is also quite determinate and can be captured more specifically by another label: fundamental ontology. Heidegger makes these points about anthropology not to disagree with Kant’s line of inquiry but to disagree with the label that Kant at one time attaches to it. Heidegger’s discussion, then, ultimately serves to erase a superficial difference between Kant’s inquiry and Heidegger’s own.

Thus, I submit, Heidegger pursues Kant’s inquiry, in line with the reconstructive interpretational style of Gadamer and Davidson. It should not surprise us that Heidegger takes up a method of interpretation that Gadamer endorses; after all, Gadamer’s interpretive method is heavily indebted to Heidegger’s work. However, what differentiates Heidegger from Gadamer is that Heidegger disagrees with Kant as he interprets him, attributing errors to Kant. Heidegger identifies and rejects an ongoing strand of argument in the Critique of Pure Reason. Indeed, Heidegger claims to find a deep inconsistency in Kant’s Critique of Pure Reason: two ongoing strands of argument that are in tension with one another. The following passage fills in the players populating his two-strand interpretation:

the power of imagination and understanding battle with each other for priority as the basic source of cognition. The battle surges back and forth, without a clear outcome. This makes the task of interpretation more difficult.

(PIK 198)

This passage paints the first Critique as a battle between two faculties for priority; which faculty is the “basic source of cognition”? The first strand of argument suggests that the faculty of imagination, the ability to represent what is not there, has priority; the second strand of argument suggests that the understanding, the ability to organize sensible information with concepts or rules, has priority. In his interpretation, Heidegger opts for the first strand, where the imagination is primary, and rejects the second strand. Heidegger, then, agrees with one strand of argument in the first Critique, but disagrees with the other strand.¹²

In contrast to a reconstructive interpretation seeking only agreement between interpreter and author, Heidegger commits himself rather to investigating the tensions in Kant’s text. Heidegger claims that “with Kant it is always worth our while actually to pursue even and precisely his labyrinths” (PIK 210). Heidegger not only acknowledges the special difficulties and inconsistencies in Kant’s thought; they are “precisely” the center of the interpretation. He attends primarily to those moments where Kant “vacillates” and “hesitates,” and even where he “unhinges himself and undermines his own foundation” (KPM 39; PIK 219–220, 279, 145–147). As Heidegger clarifies in a discussion of the first edition of the Transcendental Deduction (i.e., the A-Deduction), inconsistencies bring out the central issues with which an author struggles – that which has not yet been “settled” (PIK 220) – providing glimpses of the author’s innovative insight into the subject matter under investigation. For example, Heidegger reports that he focuses on interpreting the A-Deduction over the B-Deduction because the former “shows far more unclarity of direction and animation and a far more concrete proximity to phenomena” (PIK 220).¹³ When an author vacillates, Heidegger suggests, the author deals directly with the phenomena she is attempting to explain, struggling to bring “the darkness of the phenomena” to light (PIK 220). Pursuing labyrinths will bring out the innovations of the author’s account, so long as one isolates her insight into the subject matter, pulling it apart from misstep and error.

Heidegger does not guarantee that every philosophical text will contain labyrinths, but rather suggests that labyrinths are the mark of a “productive thinker” (PIK 220). Aristotle is his only other example of such a thinker. While “second-rate individuals perch comfortably and self-satisfied and conceited in their own opinion,” Kant and Aristotle refused to be so easily satisfied (PIK 147). Both thinkers hesitated and vacillated, experiencing “many upsets” in their pursuit of “a fundamental and radical goal” – for Kant, a fundamental ontology.¹⁴ Therefore, Heidegger suggests that inconsistency is “by no means a deficiency of [Kant’s] philosophical research.” In fact, inconsistencies are what is most “productive and instructive” in Kant’s work.¹⁵

In contradistinction to Gadamer and Davidson, Heidegger suggests that investigating inconsistencies, rather than establishing widespread agreement, serves the aim of reconstructive interpretation: finding a plausible answer to a question pursued by both author and interpreter. Though Heidegger cannot guarantee that we will discover textual inconsistencies as we interpret (as we might be reading a “second-rate” text), he offers a fundamentally different sensibility about how to receive those inconsistencies, when we do encounter them. Rather than viewing them as an indication of failure on the part of the interpreter, Heidegger sees inconsistencies as evidence of the author’s struggle to settle some philosophical issue. Heidegger, then, does not assign to the interpreter the task of eliminating, minimizing, or erasing inconsistencies. Labyrinths are the fertile ground from which philosophical innovation grows; they are a starting point from which an interpreter can begin to piece together the author’s most plausible account of the subject matter.

Thus, Heidegger demonstrates that one can interpret reconstructively while holding open the possibility of error, avoiding hermeneutic ventriloquism. Disagreement helps the interpreter locate the author’s deepest insights: the parts of the text about which the interpreter is ambivalent (agreeing with some claims, but disagreeing with others) are precisely the places where the author attempts to offer new insight into the subject matter. Further, disagreement helps the interpreter isolate the author’s best account, pulling it apart from those claims with which the interpreter disagrees. Disagreement, in sum, allows the interpreter to locate and isolate the account with which she agrees. Because Heidegger substantively disagrees with an ongoing strand of argument in Kant’s text, he cannot simply be reading himself into Kant.

2 Attributing Error to Preserve Kant’s Insight

Heidegger’s interpretation of Kant illustrates how and when it is appropriate to attribute error when interpreting reconstructively. Heidegger’s procedure in interpreting Kant suggests that we must privilege the most plausible, insightful answer to the question posed in the text; it is appropriate to attribute error as a consequence of pursuing the most promising line of response to Kant’s question. Heidegger is willing to sacrifice consistency, then, for the sake of finding the right answer.

In particular, Heidegger argues that Kant’s depiction of the imagination contains a deep insight. To preserve this insight, Heidegger must reject a great many others of Kant’s claims that are at odds with this insight. Heidegger’s discussion of each of the major sections of the first Critique takes this form. Interpreting the Metaphysical Deduction, the Transcendental Deduction, and the Schematism, Heidegger pulls apart a strand of argumentation prioritizing the imagination, and one prioritizing the understanding. He argues in each case that the strand of argument prioritizing the imagination better meets the ambitions of the section, as well as better serving Kant’s overarching goal to explain the possibility of synthetic judgment a priori.

I will illustrate Heidegger’s method of interpreting de traditione through an example: his treatment of the Metaphysical Deduction, where Kant presents the origin of the categories. The categories are the fundamental, a priori concepts that enable and structure our experience; we must possess such concepts in order to make synthetic judgments a priori (i.e., to judge about objects prior to or independently of experience). Thus, the Metaphysical Deduction is crucial for answering Kant’s leading question about the possibility of synthetic judgment a priori.

Heidegger argues that Kant’s discussion in the Metaphysical Deduction offers two competing sources of the categories; it is one battlefield between the imagination and the understanding. The more obvious source of the categories is the understanding. After all, the Metaphysical Deduction is a component of the Transcendental Analytic, where Kant has promised “an analysis of the faculty of understanding” (A65/B90). Further, Kant’s analysis in this section intends to “research the possibility of a priori concepts by seeking them only in the understanding as their birthplace” (A66/B90). According to this strand of argument, the table of categories is derived from a table of judgments offering the logical rules of thought.

Prior to Heidegger, this argument had been called into question. Heidegger cites Hermann Lotze’s challenge to Kant’s taxonomy of logical judgments, findings that seemed to doom both the table of judgments and the table of categories derived from it (PIK 178). In brief, and as William R. Woodward notes, Lotze proposed several revisions to Kant’s table of judgments in his Logik, including suggesting that all judgments are affirmative (contra Kant’s claim that three different forms of judgment, including negative judgments, fall under the heading “Quality”) (Woodward 2015: 161–162). These objections, Heidegger indicates, were widely embraced by the Neo-Kantians, resulting in widespread criticism of the table of judgments (PIK 178). However, rather than rejecting the Metaphysical Deduction with Lotze, Heidegger argues that Kant’s discussion suggests another source of the categories: the imagination. In this section of the Critique ostensibly devoted to the understanding, Kant suddenly introduces a third faculty – the imagination – just prior to offering his table of categories (A78/B103; PIK 188). If one attends to the discussion “between the table of judgment and the table of categories,” Heidegger argues, one can avoid the “all too hurried critique of the table of judgment” and locate a more promising source of the categories (PIK 179). If the source of the categories is the imagination and not the understanding, then Lotze’s objections can be avoided.

Heidegger also appeals to Kant’s theoretical machinery – Kant’s characterization of the faculties, and the characterization of the categories that is developed more fully in the Transcendental Deduction – to argue that the imagination is a more promising source of the categories. In particular, Heidegger notes that for synthetic judgment a priori to be possible, it is crucial that the categories are binding. We can draw a priori conclusions about our experience because a set of fixed, stable rules apply to that experience. Heidegger argues that a spontaneous faculty, as Kant figures the understanding, is not capable of producing binding rules. A completely active faculty can create or produce, but it cannot bind. Only a spontaneous and receptive faculty – the imagination – can both produce a rule and receive it as binding (KPM 108). Elsewhere, Heidegger suggests that this spontaneously receptive imagination is identical with care, the structure of Dasein (PIK 232). Yet at this point in the argument, Heidegger relies on Kant’s own depiction of the cognitive faculties to argue that only an origin in the imagination can secure a feature of the categories that is crucial for the possibility of synthetic judgment a priori: their bindingness. To secure this feature, we must reject the strand of argument suggesting that the understanding is the source of the categories.

Heidegger’s interpretation of the Metaphysical Deduction offers a paradigm of his method. Heidegger inquires into the source of the categories, which will help explain the possibility of synthetic judgment a priori. He finds two answers – the understanding and the imagination – and he supports the strand of argument putting forth the imagination as philosophically superior, specifically as a response to Kant’s own question and in light of Kant’s theoretical machinery. From this example, we can see that Heidegger endorses attributing error when a claim or set of claims is in tension with the most compelling line of response to the question posed by a text.

3 Heidegger’s Theory of Error

While the leading proponents of reconstructive interpretation seek widespread agreement with the authors whom they interpret, it is debatable to what extent this goal characterizes the practice of contemporary historians of philosophy and scholars of Kant in particular. In contemporary interpretations of Kant, it is common to differentiate between conflicting passages that Kant offers, and opt for one account over the other for philosophical reasons, as Heidegger does.¹⁶ While Heidegger’s interpretation of Kant resembles this contemporary approach, I suggest that Heidegger offers a more rigorous method of attributing error; in fact, Heidegger offers a novel way to meet Gadamer’s demand to treat the whole of the text. While contemporary historians of philosophy often leave aside the implausible account as soon as they motivate the plausible one, Heidegger supports his attributions of error with a theory of error. Heidegger explains why Kant errs, rather than simply discarding the strand of argument prioritizing the understanding as philosophically inferior. Heidegger explains how the first Critique, as a whole, shows Kant grappling with his insight into the constitution of the human knower, with the promising strand of argument articulating that insight, and the erroneous strand of argument retreating from it out of anxiety. Explaining Kant’s errors, Heidegger treats the text as a whole.

While Gadamer considers attributing error to be a failure of interpretation, Gadamer also endorses explaining error when this worst-case scenario cannot be avoided. In particular, Gadamer suggests that when we cannot agree with an author, we ought to explain the author’s error by appealing to psychological or historical factors:

Just as the recipient of a letter understands the news that it contains and first sees things with the eyes of the person who wrote the letter – i.e. considers what he writes as true, and is not trying to understand the writer’s peculiar opinions as such – so also do we understand traditionary texts on the basis of expectations of meaning drawn from our own prior relationship to the subject matter.… It is only when the attempt to accept what is said as true fails that we try to “understand” the text, psychologically or historically, as another’s opinion.

(TM 255–256)

At first, we read the text assuming its truth, interpreting the claims in light of our own familiarity with the subject matter. However, if we are unable to accept some claim or some subset of claims as true, despite our best efforts at charity, we retreat to a third-personal account of the author’s “peculiar opinions,” taking the error to be the result of psychological or historical barriers. This explanation represents a failure of interpretation, as one is no longer pursuing the goal of interpretation, according to Gadamer – reaching an understanding with the author, by “accept[ing] what is said as true.” The author’s “peculiar opinions” falls outside the whole of meaning (where the complete text is right about its subject matter) that the interpreter attempts to discern.

Like Gadamer, Heidegger seeks to explain an author’s errors. However, his method of explanation differs from that of Gadamer in that he does not primarily appeal to psychological or historical barriers. Further, in providing an in-depth and ongoing explanation for Kant’s erroneous strand of argument, Heidegger offers a way to account for the whole of the text that departs from Gadamer’s method. For Gadamer, accounting for the whole text means taking it to have a single, coherent meaning with which the interpreter can agree. While Heidegger, investigator of inconsistencies, does not demand such coherence, Heidegger accounts for the whole of the text because he does not discard the philosophically inferior argument, or confine his attention to the components of Kant’s discussion with which he agrees. Rather, Heidegger explains why Kant’s inquiry into the constitution of the human being resulted both in insight and in error. Heidegger provides a way to treat the whole of the text without taking the whole text to be true.

Heidegger suggests that Kant errs, because he inherits a bias common to the philosophical tradition – the predominance of logic and the understanding. The strand of argument prioritizing the imagination breaks with this bias, showing Kant’s radical and innovative insight into the subject matter. However, Kant is pulled back into the traditional bias, especially when he attempts to engage with previous thinkers.¹⁷

Historical bias, however, does not exhaust Heidegger’s explanation of Kant’s errors.¹⁸ Consider, for example, the following passage:

Kant retreats [zurückweichen] before the consequence of eliminating the priority of transcendental apperception, of understanding, that is, of the traditional, unfounded privileged position of logic. Kant is afraid [scheut sich] of sacrificing transcendental apperception to the transcendental power of the imagination.

(PIK 279)

According to this passage, Kant did not pursue his own insights into the imagination to their conclusion, because he was afraid of eliminating “the traditional, unfounded privileged position of logic.” Kant’s attachment to the traditional view certainly plays a role explaining Kant’s less compelling line of argument, but not why Kant would have retreated to that view. I argue rather that anxiety – the core of Heidegger’s theory of error – explains this retreat.

Before going any further, however, a note about vocabulary is in order. In Being and Time, Heidegger draws a distinction between anxiety (Angst) and fear (Furcht). One fears concrete, worldly happenings or things, but one is anxious about oneself – more precisely, about the kind of being that one is, about the ungrounded character of one’s own existence. Since they have distinct objects, the moods themselves are distinct. The reader might worry that Heidegger’s language in the passage above points to fear more so than anxiety: when we are afraid (as the translation puts it), we feel fear. However, in this and related passages, Heidegger does not use the term “fear” and does not use the verb that he pairs with it, fürchten (to fear). Rather, when Heidegger discusses Kant’s retreat, he uses the verbs scheuen (to be afraid or shy away) and schrecken (to frighten or horrify). Heidegger does not use schrecken when discussing either mood in Being and Time, though he does refer to Scheu (shyness) once, listing it as a type of fear (BT 142). However, Heidegger later comes to associate both terms with anxiety, defining anxiety as Schrecken (horror), and identifying Scheu (shyness) as its accompaniment. For example, in his 1935 lecture course, Introduction to Metaphysics, he refers to the uncanny,¹⁹ which “induces panicked horror [Schrecken], true anxiety [wahre Angst], as well as collected, inwardly reverberating shyness [Scheu]” (IM 166, translation modified). Likewise, Heidegger remarks on the proximity between these terms in the 1943 postscript to What Is Metaphysics?: “close by essential anxiety [wesenhaften Angst] as the horror of the abyss [Shrecken des Abgrundes] dwells shyness [Scheu]” (PWM 234, translation modified).²⁰ We will see shortly that in Heidegger’s interpretation of Kant, he also suggests that Kant retreats from, being frightened by, the abyss (KPM 117–118). At any rate, given the connection that both verbs come to have to anxiety, a single instance of Scheu in Being and Time should not decide that Heidegger in the Kant book is discussing fear.

Moreover, Heidegger draws on the verb zurückweichen (to retreat or shrink back) repeatedly in his discussions of Kant (BT 23; PIK 279; KPM 118), a term that is explicitly ambivalent between fear and anxiety. Indeed, when Heidegger differentiates between fear and anxiety, he identifies two kinds of “shrinking back” (Zurückweichen). First, there is shrinking back in this sense of fleeing; this is grounded in fear, and shrinks back from “a detrimental entity within-the-world” (BT 185). Second, there is shrinking back in the sense of falling: this is grounded in anxiety, and shrinks back from something that “has the same kind of Being as the one who shrinks back” – “Dasein itself” (BT 185). Thus, Heidegger’s claim that Kant shrinks back is ambiguous between whether he attributes fear or anxiety to Kant. Luckily, Heidegger’s distinction between fear and anxiety points to another, more substantive way to decide the issue: we can determine what mood Heidegger attributes to Kant by determining what Kant was afraid of, in Heidegger’s view. If he shrinks from some worldly entity or happening, Kant fears; if he shrinks, rather, from the kind of being that we ourselves are, then he is anxious.

At first glance, the passage on Kant’s retreat might seem to suggest a worldly event or happening that Kant feared: Kant was afraid of upsetting previous philosophical dogma (perhaps, facing rejection from those peers who subscribe to it). Indeed, this is how Henrich reads Heidegger’s claim, and he criticizes Heidegger for charging Kant with a “lack of intellectual courage” (Henrich 1994: 44).²¹ If Heidegger offered such an explanation, it would resemble the Gadamerian method of explaining error that I mentioned above, by appeal to psychological or historical factors. Kant offered an erroneous strand of argument, on this reading, due to, say, an idiosyncratic conformism. Such an explanation seems deeply uncharitable (as Henrich rightly points out), as it saddles Kant not only with error but also with a character flaw. Nor is this a convincing explanation; Kant is not a timid conformist, but the thinker who began the Critique of Pure Reason with a sharp attack on dogmatism (Aix).

I argue that the above passage on Kant’s retreat (PIK 279) gestures rather at another explanation. Kant retreated to the traditional position, prioritizing the understanding, due to anxiety: an anxiety about the very structure of the human being, rather than anxiety about upending traditional philosophical views. I argue, then, that Heidegger’s account of anxiety from Being and Time figures in his interpretation of Kant as a theory of error. Kant’s anxiety is offered as an explanation both for the erroneous strand of argument and for the inconsistency of Kant’s account (where an erroneous argument is offered alongside an insightful one). I argue, then, that Heidegger’s most controversial comments concerning his interpretive method – that his own approach is violent, that Kant takes certain positions out of being afraid or frightened – have far deeper philosophical meaning and greater interpretive merit than one might suspect.

While Heidegger’s claim that Kant is afraid might appear initially as a personal attack, Heidegger in fact argues that any ontological inquiry provokes anxiety, as well as the impulse to quell that anxiety (BT 311–312). The source of anxiety, according to Heidegger, is the constitution of the human being. For that reason, inquiring into the ontology of the human being provokes anxiety, as well as the pull to turn away from and cover over the source of that anxiety.²² Therefore, Heidegger’s theory of error does not attribute a psychological idiosyncrasy to Kant, as on the Gadamerian approach. Kant does not experience anxiety because he is an especially timid sort of person, but rather because anyone pursuing this sort of inquiry experiences anxiety.

At this point, Heidegger’s approach to interpretation, which inquires into the impulses behind some of Kant’s claims, may begin to sound like a hermeneutics of suspicion.²³ A hermeneutics of suspicion approaches the surface-level claims in a text not as truths but as symptoms of a deeper drama; with Nietzsche, one might say that the author’s claims are an expression of bodily states or physiological drives; with Freud, the claims might express unconscious desires; with Marx, they might express one’s socioeconomic position (Ricoeur 1978). Is Heidegger promoting a suspicious reading of Kant, by suggesting that his claims are expressions of anxiety?

While I will suggest that Heidegger offers Kant’s anxiety as a deep motivator of many of Kant’s overt claims, Heidegger’s commitment to reconstructive interpretation differentiates his method, as significantly more charitable and truth-oriented, from a hermeneutics of suspicion. Centrally, Heidegger is himself pursuing Kant’s question about the possibility of synthetic a priori knowledge – and Kant’s way of answering that question, by inquiring about the constitution of the human being. Whereas the deep reading of the hermeneut of suspicion is initiated by a question that the interpreter brings to the text (e.g., what physiological state do these overt moral claims reveal?), Heidegger is suspicious in the service of Kant’s question. Heidegger sees anxiety as a threat to and consequent upon Kant’s own line of inquiry – a threat to fundamental ontology – and he appeals to a deep reading when Kant’s overt argument fails, in turning away from the most promising line of response to his leading question.

Heidegger’s commitment to reconstructing Kant’s best line of argument also differentiates Heidegger from another nearby interpretive method: Derridean deconstruction. As with Gadamer, we should not be surprised if there are some affinities between Heidegger’s interpretive method and that of Derrida, as Heidegger’s method influenced Derrida. In particular, we hear echoes of Heidegger in Derrida’s insistence that texts “oscillate,” offering multiple lines of argument in tension with one another.²⁴ However, Derrida focuses only on this tension, rather than deciding on one line of argument over another. As Andrew Cutrofello puts it, Derrida thinks texts are “undecidably equivocal” (1990: 158f.) – there is no way of finally deciding the correct strand of argument, so the deconstructive interpreter is tasked with spelling out how and where texts undermine themselves. By contrast, Heidegger finds texts decidably equivocal; though there are multiple strands of argument, the best strand of argument can be identified. It can be decided, in particular, by toggling between the author’s question and the phenomena it asks about. Some strand of argument offers a better answer to that question, a better account of the phenomena under consideration. Conversely, Heidegger also wants to argue that the other strand of argument is erroneous, and explain why the author fell into this error. Heidegger steps beyond the Derridean deconstruction that notices tensions when he evaluates those strands: opting for one strand of argument, and explaining why the erroneous strand appears nonetheless.

I will now turn to Heidegger’s discussions of anxiety to fill out the details of this explanation. In Being and Time, Heidegger suggests that human beings experience anxiety due to our fundamental existential structure, whereby we must reveal the world in light of some project but have no overriding reason to pursue that project over another. We are “Being-the-basis of a nullity,” responsible for the project we do pursue, as well as those that we do not (BT 285). Due to our fundamental responsibility for the way that we understand the world, we experience anxiety. As Heidegger reviews at the end of Kant and the Problem of Metaphysics, there are two possible responses to anxiety. First, we can face that anxiety, “remembering” (KPM 164) – remembering the source of our understanding of the world. In so doing, we discover our fundamental responsibility. Second, we can flee from that anxiety, so as not to face our fundamental responsibility – “forgetfulness” (KPM 163), forgetting that and how we are revealing the world around us.

A number of passages indicate that Heidegger thinks that Kant experienced anxiety during the course of his inquiry, and that Kant responded to that anxiety by attempting to flee or forget this anxiety. These passages suggest that Kant retreats to an argument prioritizing the understanding, because he is afraid of his own insight into the imagination – “the dimension of human Dasein, into which Kant in fact looked, only to be scared away from it [zurückzuschrecken]” (PIK 189).

In particular, Heidegger suggests that Kant was afraid of assigning the imagination primacy among the faculties for human cognition, because of certain practical commitments that Kant was not willing to relinquish. In particular, Heidegger refers to the moral system that Kant outlined in the Groundwork of the Metaphysics of Morals in 1785 (between the first and second editions of the Critique of Pure Reason). While the imagination is (by Heidegger’s lights) the only faculty capable of binding us to rules, these rules are not inscribed into the imagination; the structure of the imagination is underdetermined with regard to the rules that it takes up. Rather than being governed by determinate rules, the imagination is an “abyss” (KPM 117): “in the radicalism of his questions, Kant brought the ‘possibility’ of metaphysics to this abyss [Abgrund]. He saw the unknown. He has to shrink back [zurückweichen]. It was not just that the transcendental power of imagination frightened [schreckte] him, but rather in between [the two editions] pure reason drew him increasingly under its spell” (KPM 118).²⁵ This passage suggests not only that Kant shrank back from the imagination out of fright but also that he was drawn under the spell of pure reason. In Heidegger’s view, Kant is drawn under this spell because pure reason, as opposed to the underdetermined imagination, could guarantee the stable moral rules to which Kant was committed.

Heidegger’s discussion of anxiety at the end of the Kant book identifies the argument that Kant used to quell the anxieties provoked by the abyss of the imagination. Kant offers the concrete determinations of the categories in place of fully identifying the null structure of human existence: these categories identify the features that Kant took to be the irrefutable constitution of objects. Kant thus suggests that we are bound to a single interpretive framework, rather than recognizing “the constant although mostly concealed trembling of all that exists” (KPM 167, translation modified); on this view, there is no fundamental responsibility for the way one understands the world. Kant then flees from his insight into the imagination, and offers an implausible strand of argument that portrays the categories with which we make sense of the world as the immutable rules of our understanding.²⁶

Recognizing the role of anxiety in Heidegger’s interpretation of Kant provides us with a better, even transformative understanding of Heidegger’s famous admission that his interpretation is “violent” (PIK 247). In his account of anxiety in Being and Time, Heidegger claims that “existential analysis … constantly has the character of doing violence, whether to the claims of the everyday interpretation or to its complacency and its tranquillized obliviousness” (BT 311). Heidegger suggests that everyday interpretation flees from anxiety by “closing … off” interpretation (BT 311), taking its understanding of the world to be the only one possible. Existential analysis does violence to this everyday interpretation by undermining it, revealing that we are bound neither to a single set of categories nor to a definitive list of moral rules. The professed violence of Heidegger’s interpretation is not his confession to twisting Kant to his own purposes, as many commentators charge; rather, Heidegger’s interpretation is violent in that it undermines the “everyday” explanation that Kant offered in order to quell his anxiety. However, this violence leaves untouched the strand of argument that genuinely pursues the phenomena: the strand of argument prioritizing the imagination.²⁷

My account of Heidegger’s interpretive violence is also consonant with Heidegger’s discussion of destruction in Being and Time, where he calls for a destruction of the philosophical tradition (with the Kant book being his first extended implementation thereof). In Being and Time, Heidegger emphasizes that his interpretive method has a positive component and a negative component. It destroys those philosophical elements that obscure rather than offer genuine insight into the phenomena under discussion; thereby, the interpreter retrieves the genuine insights that past philosophy has to offer (BT 22–23). In the context of the Kant interpretation, Heidegger rejects the strand of argument prioritizing the understanding in order to retrieve genuine insight into the ontological constitution of Dasein, and thus to achieve the goal of fundamental ontology. Echoing Kant’s own wording in the Prolegomena, Heidegger suggests that a “resolute” reader is required to follow Kant’s arguments to their appropriate conclusions, rather than retreat from them with Kant (Ak. 4:274; KPM 170). Heidegger’s theory of error offers a general recommendation: when we see an author struggling with the sorts of questions with which Kant struggles, we must seek the tensions and the labyrinths, and be on guard against easy and comforting explanations.

4 Heidegger’s Later Retraction of the Kant Book

Heidegger’s own critical comments on his late 1920s interpretation of Kant may seem to weigh against my view of Heidegger’s interpretive method and its characteristic violence. Indeed, commentators often appeal to these comments in order to support the view that Heidegger is a hermeneutic ventriloquist in his late 1920s interpretation of Kant.²⁸ Arguably, the most worrying of these comments are offered in his prefaces to the second and fourth editions of Kant and the Problem of Metaphysics, published in 1950 and 1973, respectively.²⁹ Before working through these comments, there are two issues to disambiguate: first, the method that Heidegger used in interpreting Kant in the late 1920s, which I have reconstructed and defended in this chapter; second, the way that Heidegger carried out or executed this method. I will suggest that Heidegger’s later comments offer very little reason to reject the interpretive method that I have defended in this chapter, and that they offer unpersuasive reasons for thinking that he carried out that method incorrectly.

In the 1950 preface to the second edition, Heidegger seems to criticize his late 1920s method of interpreting Kant, particularly its violence. As I have outlined above, Heidegger initially identifies violence as a feature of his interpretive method: violence is to be used, in textual interpretation (KPM 141) and existential analysis more broadly (BT 311), against those errors that obscure deeper insights. However, in the 1950 preface, Heidegger begins speaking of his violence not as a feature but as a bug of his method. This discussion is worth quoting at length:

Readers have taken constant offense to the violence of my interpretations. Their allegation of violence can indeed be supported by this text.

Philosophicohistorical research is always correctly subject to this charge whenever it is directed against attempts to set in motion a thoughtful dialogue between thinkers. In contrast to the methods of historical philology, which has its own agenda, a thoughtful dialogue is bound by other laws – laws which are more easily violated. In a dialogue the possibility of going astray is more threatening, the shortcomings more frequent.

(xx)

In 1950, Heidegger portrays violence as a sort of mistake into which an interpreter can easily fall when interpreting a text, a mistake that becomes possible when that interpretation attempts a thoughtful dialogue instead of a philological or scholarly interpretation. This violence amounts to a tendency to go overboard, which works “against” having a thoughtful dialogue with another thinker. Heidegger’s later depiction of his interpretive violence seems to capitulate to his critics, rather than capturing the actual role of violence in the interpretation; he depicts his violence as a shortcoming, rather than an interpretive tool used to clear away error. These later comments do not do justice to his interpretive method in the late 1920s. Because Heidegger equivocates on the term “violence,” his comments provide little reason to reject the violence that he initially builds into his interpretive method – the violence that clears away unconvincing, anxiety-driven arguments, so that one can get to the truth about the subject matter.

The precise mistake to which Heidegger admits in this passage speaks to the second issue – his success in carrying out his interpretive method. Which “law” of interpretation did Heidegger “violate,” in his later estimation; in what way did he “go astray”? Heidegger’s comments in the preface to the fourth edition shed light on this issue: Heidegger suggests that he “overinterpret[ed]” Kant by interpreting “the Critique of Pure Reason from within the horizon of the manner of questioning set forth in Being and Time” though “in truth … Kant’s question is foreign to it” (KPM xviii). The problem, in short, is that he got Kant’s question wrong, taking it to be similar to the one posed in Being and Time. Heidegger suggests that his later interpretations of Kant attempt to correct this error – “to retract the overinterpretation” (KPM xviii).

In the 1935–1936 reinterpretation of the Critique of Pure Reason, he attributes a different question to Kant, marking the space between Kant’s question and his own. I would argue that the reinterpretation is in fact weaker than the late 1920s interpretation of Kant. The new question that Heidegger attributes to Kant – namely, what is a thing? (WT 37) – fails to do justice to Kant’s Copernican revolution, where Kant turns his attention from “the objects” to “the cognition” that makes them possible (Bxvi). Kant asks not only about things but also about the human knower who experiences these things, as recognized in the late 1920s characterization of Kant’s project as a fundamental ontology.³⁰

While I think there are reasons to object to Heidegger’s later criticism of his own interpretation of Kant, it is important to recognize that the issue at hand – whether Heidegger accurately identified Kant’s question – does not undermine the interpretive method that I have defended in this chapter. In fact, Heidegger’s later concerns with the interpretation show a continued commitment to this interpretive method. Specifically, Heidegger expresses his continued commitment to a central tenet of his late 1920s interpretive method: the interpreter must pursue the question posed by the author. And it is through adhering to the tenets of this interpretive method – pursuing the question posed by the author, taking up the concepts offered to answer that question, identifying the most plausible line of response, and identifying and explaining the errors that obscure that plausible line of response – that an interpreter avoids hermeneutic ventriloquism.

5 Conclusion

While many see Heidegger as reading himself into Kant, I have argued that Heidegger’s distinct approach to reconstructive interpretation avoids this mistake. Heidegger keeps fixed the leading question of Kant’s analysis, takes on Kant’s theoretical machinery, and identifies the most plausible line of response. However, unlike some proponents of reconstructive interpretation, he also attributes error; he identifies and rejects an implausible line of argument appearing in Kant’s text. Heidegger’s theory of error, further, explains why Kant went astray. Heidegger thereby demonstrates how one can interpret reconstructively while avoiding the risk of hermeneutic ventriloquism.

In interpreting Kant as he does, Heidegger attributes to Kant a compelling, innovative insight into the imagination: the categories of thought originate in the imagination rather than the understanding. While this interpretive method may not read a work as offering a consistent, and consistently convincing, line of argument, it offers its own kind of charity: it seeks out and attempts to foster the growth of the best, most exciting ideas that one encounters in what one reads.