The Myth of the Taken: Why Hegel Is Not a Conceptualist

ABSTRACT

The close connection often cited between Hegel and Wilfrid Sellars is not only said to lie in their common negative challenges to the ‘framework of givenness,’ but also in the positive lesson drawn from these challenges. In particular, the critique of givenness is thought to lead to a conceptualist view of perceptual experience. In this essay, I challenge the common idea that Hegel’s critique of givenness provides specific support for a conceptualist view. The notion that Hegel, if anyone, is a conceptualist depends on faulty assumptions about the conceptual character of all language, including the indexical expressions Hegel discusses in ‘Sense-Certainty.’ I first show that these assumptions are often imported into Hegel’s texts but are also out of keeping with his own systematic views of concepts and language. To avoid a merely verbal disagreement, however, I then explore the features of Sellarsian semantics needed to make a thorough conceptualism plausible. Sellars’ ‘picturing’ theory is necessary to show how non-predicate terms (like indexicals) have meaning, but in order to put this feature of Sellars’ semantics in service of a conceptualist view, one must abandon the descriptive character of concepts that is a minimal feature of Hegelian thought.

KEYWORDS:

• G.W.F. Hegel
• Wilfrid Sellars
• Myth of the Given
• conceptualism about experience
• perception

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. This description matches quite well the view McDowell attributes to Sellars: ‘[E]xperiences are actualizations of conceptual capacities in sensory consciousness’ (2003, 77). Here, conceptual capacities figure in a real definition of experience. Note that Sellars discusses the older (and largely unrelated) sense of ‘conceptualism’ at EPM § 25. See also Hanna (2013).

2. Ironically, though Sellars’ influence is responsible for inspiring contemporary conceptualism, especially McDowell’s, it is sometimes unclear in what sense his own views should be thus characterized. For in his major work after EPM, Science and Metaphysics (1968), Sellars spoke of conceptuality being ‘guided’ by pure receptivity of a non-conceptual variety. McDowell criticizes Sellars on this point in his Woodbridge Lectures (reprinted in McDowell 2009a). However, Levine (2016) argues that the Sellarsian non-conceptual content is not intentional in a way that threatens his conceptualism. Though I will speak of conceptualism in a fairly general way in the early portions of this paper, I will consider what a thorough Sellarsian conceptualism would look like in section 3.

3. This is how I would characterize Robert Pippin’s claim here: ‘This [sc. Hegel’s position] can fairly be called an idealism since it seems to make the possibility of experience, experiential knowledge, and explanatory success dependent on conceptual rules that are not themselves empirically derived, given that the possibility of empirical experience already depends on such discriminating capacities’ (2005, 383).

4. ‘For it is with the denial that a firm distinction can ever be usefully drawn between intuitional and conceptual elements in knowledge that distinctly Hegelian idealism begins, and Hegel begins to take his peculiar flight, with language about the complete autonomy, even freedom of “thoughts” self-determination’ and “self-actualization”’ (Pippin 1989, 9). Sellars anticipated, but tried to avoid, a similar path: ‘Indeed, it is only if Kant distinguishes the radically nonconceptual character of sense from the conceptual character of the synthesis of apprehension in intuition [which is, of course, to be distinguished from the conceptual synthesis of recognition in a concept, in which the concept occupies a predicative position] and, accordingly, the receptivity of sense from the guidedness of intuition that he can avoid the dialectic which leads from Hegel’s Phenomenology to nineteenth-century idealism’ (1968, § 1.40).

5. Cf. Sellars (1968, § 1.17).

6. Hegel’s works cited in text will reference the Werke German edition(Hegel 1970), an English translation (when available), then a paragraph number for the Encyclopedia or for Terry Pinkard’s translation of the PhG.

7. From the other side: ‘It is clear that though he does not appeal to specific texts, Sellars takes his campaign against the Myth of the Given to be Hegelian in spirit’ (McDowell 2003, 76).

8. Other references include Pippin 1997, 10–11; Westphal 2000, 175; Rorty 2003, 42; Stekeler-Weithofer 2005, 239; Rockmore 2005, 62; de Vries 2008, 65 et passim; Maher 2012, 124, n. 21; Hanna 2013, 3. Selivanov (2012) offers one of the more thorough comparisons. Bowman (2012, 105 n. 58; 106) sounds a note of caution at the connection.

9. See also Brinkmann (2011, 108–117). Brinkmann’s reading exemplifies a problem we will return to later on, namely the identification of deictic (or indexical) expressions with concepts (even predicates), though Brinkman acknowledges that ‘the conceptual information contained in the concept of This (or of Here, or Now) is close to zero’ (109).

10. Cf. EG § 467, addition.

11. As far as I can tell, the only uses of ‘concept’ in ‘Sense-Certainty’ are references to the meta-conception of sense-certainty, i.e., its notion of what its knowledge and putative object are: ‘That is, [the object in sense-certainty] is to be considered as to whether this, its concept, which is to be the essence, corresponds to the way it is present within that certainty’ (PG 84/61, § 94). A similar reference is found in ‘Perception’: ‘According to its simple concept, this experience can be briefly looked at in this way’ (103/76, § 126). This is a concept of not in perception.

12. Hegel says consciousness is implicitly conceptual when ‘the unconditioned universal’ is on the scene, during the later stages of ‘Perception’ (107/79, § 132), which becomes thematic in ‘Force and Understanding.’ This suggests that consciousness is not yet implicitly conceptual in ‘Sense-Certainty.’

13. I put it this way here (following Westphal (2000)) to recall that the allusion to Begreifen Hegel makes was to an attempt at avoiding conceptualizing, not doing its opposite. It is true that a negation of a genuine opposite would entail its contrary affirmation. Nothing like that relationship between conceptual and non-conceptual is explicit in the text, however; nor would it be proposed, I expect, by conceptualists.

14. His sole mention of ‘intuiting’ in the passage (‘I am pure intuiting,’ 88/64, § 104), moreover, is quite removed from the typical Kantian sense, though he is capable of using the term in roughly Kant’s way (cf. 54/74, § 80).

15. Perhaps Klaus Brinkmann (2011, 102) speaks for many commentators when he assumes that the ‘Notion/Concept’ is ‘tacitly presupposed’ in the discussion of sense-certainty.

16. According to Dina Emundts, ‘Most interpreters of Sense Certainty immediately identify the universal that is introduced in this chapter with concepts’ (2012, 179). However, if we follow her reading, this identification is premature: ‘First of all, in my view, the assumption that Hegel argues that any reference to something in the world is conceptual is not appropriate. Hegel does not want to deny to consciousness the ability to form opinions [Meinungen] about some instant sensuous presence. Only it cannot defend these opinions as true’ (2012, 180).

17. See Emundts (2012, 180f.), for the view that Hegel affirms space and time as universals here, much like Kant himself. ‘Quantitative and qualitative determinations are the universal but not yet something conceptual in the sense of something law-like [Gesetzmäßigem] (which the determines the relationship of objects among themselves). And it is also not something conceptual in the sense of a predicate-concept’ (ibid., 181).

18. Michael Forster takes it as a significant assumption on Hegel’s part (argued elsewhere, he claims) that ‘meaning and thought require linguistic expressibility (including pointing as a form of language)…’ (1998, 205ff.).

19. Cf. Sellars (1974, 1991 essays 10 and 11).

20. ‘In experience one takes in, for instance sees, that things are thus and so. That is the sort of thing one can also, for instance, judge’ (McDowell 1994, 9).

21. This does not mean that Sellars equated concepts and words, which he denied (cf. EPM § 31). They must be caught up in some rule-governed or inferential pattern. See section 3 for a further discussion.

22. Conceptualist readings such as those of McDowell and Pippin tend to assume that there is a specific problem about the way sensory knowledge is approached in ‘Sense-Certainty’ which would be avoided if the proper role of concepts was acknowledged. In my view, Hegel’s argument hangs on features of the representation of immediate spatial-temporal content that would be shared regardless of the acknowledgment of conceptuality. Bowman (2012) seems to arrive at a similar view, though from his view of Hegel’s metaphysics.

23. This is the view put forward by Charles Travis (2004), with McDowell’s conceptualism specifically in view.

24. I offer support of these views especially in Wolf (2018). I also provide developmental, historical reasons to think that Hegel uses ‘concept’ in ways coordinate with its typical philosophical meaning in Wolf (2017).

25. E.g., Bowman (2013) and Taylor (1977).

26. As Pippin (2015) also attempts (however unsuccessfully, in my view).

27. E.g., ‘But a concept is also, first of all, the concept, and this concept is only one concept, the substantial foundation…’ (WL I: 29–30/19).

28. For further references, see, e.g., WL II: 290/541–42; 321–22/564; 260/519.

29. One should note, however, that he also makes remarks of contrary nature: ‘[T]he concept as deduced here should in principle be recognized in whatever else is adduced as such a concept’ (WL II: 252/514). And: ‘[T]he deeper significance of the concept is in no way so alien to general linguistic usage as it might seem to be at first sight’ (EL 308/237, § 160Z).

30. Cf. Bowman (2013, 32); Horstmann (2017, 133–34).

31. Though I am simplifying Kant’s views here, one might think I am leaving out the crucial claim that concepts are rules, which many Sellarsians find attractive. However, in the primary context Kant discusses such rules, they are hardly rules of inference, but rather ‘schematic’ rules that order representations on the basis of their spatial and temporal properties (A 138ff./B 177ff.). For example, the ‘rule’ for a concept of a plate involves its circular geometrical shape (A 137/B 176). I find this far from an inferential rule that would give a logical shape to a concept.

32. See, e.g., WL II: 282/536, 525/717; W 7: 123; W 9: 110.

33. Cf. WL I: 48/32.

34. See Kant’s Jäsche Logik, § 1 (Kant 1992, 589; Ak. 9: 91).

35. Cf. Hegel (1991 [= EL], xlvii).

36. That is how Hegel completes the sentence just quoted: ‘to recognize the concept in what is otherwise mere representation is philosophy itself’ (WL II: 406/628).

37. Cf. EL 24/11 (1827 Preface): ‘…so, too, there are two tongues [Sprachen] for that import: the tongue of feeling, of representation, and of the thinking that nests in the finite categories and one-sided abstractions of understanding, and the tongue of the concrete Concept.’

38. The concept need not be an adequate or true understanding: the latter would fall under the category of ‘idea’ (Idee) for Hegel.

39. Thus, I am not accusing conceptualists of ‘intellectualism’ (à la Hubert Dreyfus). My own view of Hegelian concepts would lead to intellectualism if I suggested they were as involved in experience as McDowell and others suggest.

40. See Brandom’s notion of de re in contrast to de dicto belief-ascription (where the latter is closer to ‘intellectual biography’) (2002, 99–107).

41. Recall the controversy mentioned in note 2 above regarding whether such a view is Sellars’ own.

42. Sellars does briefly consider the way empiricists also attempt to credit acquaintance with sense-data considered as particulars, rather than predicative facts. However, he charges this attempt with confusion if meant to explain the way such particulars form the basis of knowledge (taken as propositional). Cf. EPM §§ 3–5.

43. Cf. Sellars (1979, 100–101).

44. For present purposes, it is not necessary to detail Sellars’ motivation for this position. Most basically, he sees this semantics as required for his ontological nominalism. Apart from his own scientific leanings toward nominalism, he saw the realism of contemporaries like Gustav Bergmann as fraught with paradox. See ‘Naming and Saying’ in his 1991.

45. The account here derives especially from Sellars (1979, Ch., 5). See also essays 2, 6, and 7 in Sellars (1991).

46. See deVries (2005, 81–89) and Rosenberg (2007, Ch., 5) for sympathetic accounts of picturing. Levine (2007) offers a helpful critical evaluation within Sellars’ attempt to conjoin the two ‘images.’

47. Sellars himself (1979, §§ 5.42ff.) develops the map analogy.

48. Note Sellars attitude toward the structure of ordinary language: ‘ordinary grammar is the paper money of wise men but the gold of fools’ (1991, 208).

49. Cf. also: ‘A map is no list of names, though in a sense it consists in names’ (1979, 112).

50. It would be wrong to say that Sellars reduces saying to naming, since he is concerned to keep the distinction relevant. However, the predicative structure of saying is not to be accepted as explanatory in its own right. Brandom expresses skepticism (rightly, in my view) that Sellars has the resources to make ‘saying’ intelligible: ‘I don’t see that we have the makings of a story [in Sellars] on the ontological side or on the semantic side of what corresponds on the pragmatic side to saying (claiming, believing) something’ (2015, 270). Thanks to an anonymous referee for pressing this point.

51. See his (1968, § 3.32) where Sellars says explicitly that such expressions are necessary for the ‘language entry rules’ which give language meaning.

52. Names and indexicals are closely related for Sellars. More specifically, indexicals function as cataphoric (or anaphoric) place-holders for names. In Rosenberg’s reconstruction of Sellars’ view, ‘Roughly, ·this· is a temporary proper name’ (2007 [orig. 1978], 149).

53. See also PhG 62/41, § 66.

54. As David Kaplan writes, ‘But in any case, the descriptive meaning of a directly referential term [sc. such as a proper name or an indexical] is no part of the propositional content’ (1989, 497). Kaplan’s point is ably illustrated by Hegel’s own claim that the demonstrative ‘now’ cannot refer to the same thing if repeated later on. To retain the same propositional content, one would have to replace it with ‘then.’

55. ‘All that would be needed for a bit of [experience] to come to constitute the content of a conceptual capacity, if it is not already the content of a conceptual capacity, is for it to be focused on and made to be the meaning of a linguistic expression. … No aspect [of experience] is unnameable, but that does not require us to pretend to make sense of an ideal position in which we have a name for every aspect, let alone to be in such a position’ (McDowell 2009b, 319–20).

56. It is true that Sellars wants us to think that ‘naming’ and other demonstrative pragmatics will function like describing in a holistic ‘world story,’ but this appears to be conjectural, especially given the weakness of the example of Jumblese for modeling natural languages. Sellars is aware of this general problem (cf. his remark on describing vs. labeling in 1957, 306), but his reliance on the holism of world stories to solve it seems to be a mere promissory note. (Thanks to an anonymous referee for requesting further clarification here.)

57. The research for this project was generously supported by a Rev. John P. Raynor, S.J., Fellowship at Marquette University. Many thanks also to Jorge Montiel and Phil Mack for helpful discussion on an earlier draft of the paper, as well as to an anonymous reviewer for valuable feedback.