Abstract
A dualism characteristic of modern philosophy is the conception of the inner and the outer as two independently intelligible domains. Wittgenstein’s attack on this dualism contains deep insights. The main insight (excavated from §304 and §293 of the Philosophical Investigations) is this: our sensory consciousness is deeply shaped by language and this shaping plays a fundamental role in the etiology of the dualism. I locate this role in the learning of a sensation-language (as described in §244), by showing that this learning is, under another aspect, the incision of language, namely the infliction of cuts upon certain natural-primitive unities between the inner and the outer. These cuts, driven by powerful forces, eventually harden into an entrenched division between the inner and the outer, thereby providing a constant soil for the dualism. That this dualism is rooted in the very learning of a language is cause for ambivalence about language.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
A classic philosopher who assumes this dualism (self-consciously or not) is Descartes, for whom the inner, the entire domain of consciousness, remains intelligible even if there is no external world at all.
The label ‘the Private Language Argument’ is in certain ways misleading. I shall use it only to designate a stretch of text, say PI, §§243–315. I shall cite passages in Part I of the PI in the style of ‘§304’ and ‘§304b’ (the second paragraph of §304) and passages in Part II of the PI in the style of ‘PI, p. 187’ or ‘PI, II.xii’.
Here I am using ‘action’ and ‘behavior’ along lines regimented by the distinction between intelligence and sentience. The same regimentation is also in force in note 5, but not generally in this essay.
See McDowell’s “One Strand in the Private Language Argument”, collected in Mind, Value, and Reality, pp. 279–296. It is true that in this essay McDowell does not regard the strand that he discusses as Kantian. In fact, he describes sensations conceived as chunks of non-conceptual given that justify conceptualizations as intuitions “in a more or less full-blown Kantian sense” (p. 285). But McDowell’s Wittgenstein is actually quite close to Kant, if we read Kant more sympathetically than he did in that essay. For a more sympathetic reading, see Stephen Engstrom’s “Understanding and Sensibility”. It should be noted that McDowell himself has shown increasing sympathy to the Kantian notion of an intuition, for example in his Mind and World (1994) and still more in “Avoiding the Myth of the Given” (2009). For Sellars’s label “the Myth of the Given”, see his Empiricism and the Philosophy of Mind.
This becomes obvious with two points. On the one hand, one cannot distinguish between the conceptual and the non-conceptual in animals without any conceptual capacity, but such animals do have both an inner and an outer life. On the other hand, the same single aspect of animal life takes different forms in different animals. Thus the inner aspect takes conceptual form (i.e. thought) in rational animals and non-conceptual form (i.e. sensation) in non-rational animals; and similarly for the outer aspect.
The phrase “logically non-transferable” is borrowed from P. F. Strawson (Individuals, pp. 95–98). It is derivable from §398, in which Wittgenstein compresses the thrust of the Private Language Argument into one single sentence: “if as a matter of logic you exclude other people’s having something, it loses even its sense to say that you have it.” (I have restored the word “even [auch]”, missing both in Anscombe’s translations and in the revised translation by Hacker and Schulte.)
This is a highly simplified picture of the dialectic between the idea of a private object and behaviorism. In particular, I have left out hybrid accounts that try to weld together a private Inner aspect and an outer aspect. Wittgenstein attacks such hybrid accounts in, e.g., §273 and §280. But these complexities can be profitably left out in this essay.
This paragraph is a brief summary of my “‘It is not a something, but not a nothing either!’ – McDowell on Wittgenstein”, which is a preparatory piece for this essay. In that piece, I argue in detail for the interpretation of the paradox summarized here (by way of correcting a misreading of §304 and §293 by McDowell).
Such readers include Robert Fogelin, Cora Diamond, and David Finkelstein.
Fogelin (1996, p. 43) says: “The clear implication of this passage [§304b] is that philosophical problems about the mental have arisen through treating mental ascriptions on the model of talk concerning chairs, houses, and the like”. He has replaced Wittgenstein’s list with his own “chairs, houses, and the like”, planting a non-existent sign-post pointing in the direction of outer objects.
Diamond (2000, p. 276) thinks the message of §304b is that “we need to make a break with the idea that language always functions in the same way: whether to convey thoughts about Bismarck’s head or his headache”. Context shows that she is contrasting outer objects with inner objects, which she takes Bismarck’s head and his headache to stand in for, respectively.
Finkelstein (2003, p. 138) thinks that §304b is urging the reader “to stop philosophizing as if our talk about pains functioned in more or less the same way as our talk about houses”. Context shows that he takes pains to stand in for inner objects and houses to stand in for outer objects.
One might object here that the intellectual function of language is broader than the function of language to convey thoughts, on the ground that one can think alone, that is, without conveying one’s thoughts to anyone else, even though one’s solitary thinking does involve making intellectual use of language (thought as “inner speech”, on analogy from overt speech). The reply is that the word translated as “convey” here, übertragen, does not connote the communication of one’s thoughts to others, as “convey” does. It would be perfectly acceptable to translate the relevant text not as “to convey thoughts” but as “to render thoughts”, in the sense of “to formulate one’s thoughts in words”, which one can do alone (silently or aloud). Wittgenstein’s phrasing is general enough for the classification “intellectual” to fit.
This is so with the abovementioned commentators: Fogelin, Diamond, and Finkelstein. That the differences they stress are still within the intellectual function of language becomes plain if we treat (in accordance with Wittgenstein’s spirit) using language to convey thoughts about things as essentially the same as talking about things. Thus:
Fogelin stresses the difference between talk about the outer objects and talk about inner objects (“mental ascriptions”),
Diamond stresses the difference between conveying thoughts about outer objects and conveying thoughts about inner objects,
Finkelstein stresses the difference between talk about outer objects and talk about inner objects.
Wittgenstein shows that he is well aware of this continuum at PI, p. 187: “Are the words ‘I am afraid’ a description of a state of mind? I say ‘I am afraid’; some else asks me: ‘What was that? A cry of fear [\(\hbox {expressive}_{\mathrm{s}}\)-affective]; or do you want to tell me how you feel [intellectual-descriptive]; or is it a reflection on your present state [intellectual-reflective]?’—Could I always give him a clear answer? Could I never give him one?” He then gives a variety of cases (PI, p. 188). It is worth noting that some mixtures of the intellectual and the \(\hbox {expressive}_{\mathrm{s}}\) functions of language are so common as to have become institutionalized in specialized senses of words. Thus, e.g., a patient is customarily said to complain of this or that to the doctor. (Note that language is not exhausted by its intellectual and \(\hbox {expressive}_{\mathrm{s}}\) functions. There are functions falling outside this classification, e.g., using someone’s name to call him, talking to a prelinguistic child to comfort it or entertain it, and doing a lot of sheer talking to improve one’s pronunciation or to acquire a desired accent.)
I have not offered a systematic treatment of the \(\hbox {expressive}_{\mathrm{s}}\) function of language here, partly because this is not needed for my purposes, partly because a systematic treatment is difficult due to the enormous variation within the above-mentioned continuum. Dorit Bar-On offers a systematic treatment of expression in Speaking My Mind (Chapt. 7), by developing a threefold distinction amongst expressions (her \(\hbox {EXP}_{1}-\hbox {EXP}_{3}\)). But for reasons I shall not elaborate here, this distinction is not suitable for characterizing the contrast I want to emphasize: the contrast between the sophisticated (the intellectual) and the natural-primitive (the sensuous, especially the affective).
This difficulty is manifested, for instance, in the common failure of commentators on §304b (discussed earlier in this section) to look away from the intellectual and toward the natural-primitive. Note that this is a difficulty that afflicts us in philosophy, not in ordinary life. In ordinary life we have no difficulty in recognizing the relevant natural unities, e.g., the unity of an “Ouch!” and the pain that it \(\hbox {expresses}_{\mathrm{s}}\). Further, this recognition is immediate, in the sense that it is not by inference or interpretation or theory-ladden observation. Rather, we directly perceive, say, the pain in the “Ouch!” As Wittgenstein reminds us, we can often directly perceive the inner states of another person in his bodily behavior, especially in his facial expressions. (See Zettel, §§220–225 and Last Writings on the Philosophy of Psychology, Vol. 1, §§735–757.) Bar-On (Speaking My Mind, pp. 270–272) develops a kindred view about natural expressions of inner states: natural expressions show certain inner states of a subject, in the sense of making them directly perceptible (to naturally suitable observers).
Note that the beetle in the box is not the concrete analogue of a sensation properly conceived, but that of a sensation conceived as a private object. This is because the box is a pocket of privacy: no one else can look into one’s own box. I discuss this in detail in “‘It is not a something, but not a nothing either!’ —McDowell on Wittgenstein”.
I said earlier (tacitly at the end of Sect. 2) that questions of right are not pressed against the idea of privacy in §293. This was a slight overstatement, since in §293a one can of course discern a question of justification pressed against the irresponsible generalization from one’s own case to the cases of all others. But this is a very minor qualification, because the main content of §293, i.e., §293b and §293c, is quite distant from that irresponsible generalization and contains no question of right.
This is a slight expansion on Wittgenstein’s “construing the grammar of the expression of sensation on the model of ‘object and designation”’. It is plain what misuse of the model is intended by him, i.e., taking the sensation to be the object and its expression to be the designation.
See McDowell, “One Strand”, pp. 284–286.
See McDowell, “One Strand”. I discuss this point in detail in “‘It is not a something, but not a nothing either!’—McDowell on Wittgenstein”.
Basically the same dynamic also drives hybrid accounts that try to combine a private Inner aspect and an outer aspect (either a behavioristic Outer aspect or an ordinary outer aspect). See note 7.
Such readers include Peter Hacker, Meredith Williams, and John Canfield (listed here in the order of decreasing explicitness in advancing the reading in question).
Hacker’s view is explicit and well documented in his Analytical Commentary, Vol. 3 (1990). See both his exegeses of §293 and §304 and his interpretative essays, for example, “Avowals and Descriptions”, Sect. 4, end of second paragraph; “Behaviour and Behaviourism”, end of Sect. 3; “Thinking: the Soul of Language”, beginning of Sect. 4. (Note that he uses ‘object and name’ instead of ‘object and designation’.)
Williams (1999, p. 33) says that “according to Wittgenstein [...] the ‘object and designation’ model [...] cannot apply to sensations. Sensations are not objects (which disappear when stripped of their accoutrements), but states of living organisms”. It is true that sensations are states of living organisms, but that is no reason to think that the ‘object and designation’ model cannot be used on them.
Canfield (2001, p. 380) says: “[...] hence his remark: ‘We have only rejected the grammar which tries to force itself on us here’ (PI, §304). That suspect grammar takes talk of pain on the model of talk of objects.” It seems fair to assume that by ‘objects’ here Canfield means physical objects, for otherwise it is unclear how the modeling is supposed to be suspect. (Canfield is here relying, implicitly but correctly, on the close connection between §304 and §293.)
This misreading of Wittgenstein’s diagnosis in §293c is essentially the same as the misreading of his diagnosis in §304b that I discussed earlier (Sect. 3, note 9). Both take Wittgenstein to be warning against assimilating inner objects (e.g. pains) to outer objects (e.g. houses), the difference being merely that here the warning is interpreted on the basis of a narrower sense of ‘object’. But the real warning is directed at something else that is much less obviously wrong than that assimilation.
This sense of Gegenstand comes out especially clearly in its negative use. I once received a letter from the German authorities informing me of the obligation to register my apartment in Berlin. It included an instruction whose gist was: “If you have already registered, please regard this letter as gegenstandslos.” The word gegenstandslos here (literally “without object” or “object-less”) can be translated as “irrelevant” or, more idiomatically, “moot”.
Anscombe’s rendering of Gegenstand as “subject” (rather than as “topic”) in these places brings out a phenomenon that she has herself drawn attention to, namely that the words “object” and “subject” have suffered a partial double reversal of meaning (“The Intentionality of Sensation”, in her Collected Philosophical Papers, Vol. 2, p. 3).
The phrase also has Strawsonian echoes. See Strawson’s Individuals, especially Part II.
Of course language as intellectual equipment also includes other crucial structures, e.g., the logical structures captured by such words as ‘not’, ‘or’, and ‘if’. But the structure of ‘object and designation’ or (what is really the same) the structure of ‘subject and predicate’ remains fundamental. Intellectual operations such as negation and inference are parasitic on predication.
Wittgenstein uses Bezeichnung and Bennenung (and their verb forms) essentially interchangeably in the PI.
This is an important (because rope-strengthening) point of overlap between Wittgenstein’s attack on the dualism of the Inner and the Outer and his attack on the dualism of conceptual scheme and non-conceptual given (Sect. 2). This overlap becomes obvious when we realize a basic fact, namely that the intellectual use of language is its predominant use in our life (civilized human life). To have a language is, first and foremost, to have an intellect or to have reason. So the above point that the sensory consciousness of us talking or rational animals is shaped by language or reason is just the Kantian point that McDowell has excellently brought out.
This claim is defended in my “A Meeting of the Conceptual and the Natural: Wittgenstein on Learning a Sensation-Language”. This paper offers a detailed interpretation of the whole of §244, in connection with a special kind of “naturalism” in Wittgenstein.
One may object to the “only”, noting that the infant, when in pain, may also wince or twist its body. But when an infant winces or twists its body in pain, this is part of its crying. Anyone who has witnessed an infant cry knows it is not a purely vocal reaction but involves the entire little body, sometimes quite violently. (Like Wittgenstein, we are here concerned only about normal cases, excluding such cases as infants born mute but sensitive to pain.)
The saying of “Ouch” in pain is similar to the primitive cry of pain in that they both have an inarticulate significance. They are also acoustically close (this is not an accident of English: exclamations of pain have a similar sound in many languages). On both scores, the contrast between “Ouch” and a proper word, e.g. “pain” in English or “Schmerz” in German, is obvious.
Wittgenstein’s original passage reads: “The origin and the primitive form of the language-game is a reaction; only from this can more complicated forms develop. [paragraph break] Language—I want to say—is a refinement, ‘in the beginning was the deed’.” (Culture and Value, p. 31) He likely has in mind both the development of a language itself and the development of a language in an individual human being. I shall restrict myself to the latter kind of development (which amounts to the learning of a language), though some formulations fit both kinds.
A great deal more would need to be said to properly elucidate these partially constitutive connections. I can only briefly touch on a few points here. First, the structuring of the inner by the outer, though salient in children, also happens in adults. In a special region of the outer, i.e., facial expression, there is some interesting field work by Jonathan Cole (“On ‘Being Faceless’: Selfhood and Facial Embodiment”) documenting how people (mostly adults) whose faces have been rendered expressionless by a medical condition sometimes suffer a corresponding impoverishment of their inner life, especially their emotional experiences. Secondly, the constitutive role of the outer for the inner is especially striking in the case of a crucial class of bodily sensations which under normal circumstances cannot be described except in terms of one’s bodily parts and positions (e.g. “bent knees”). Cf. Anscombe’s pioneering discussion of non-observational knowledge of the position of one’s limbs (e.g., Intention, §8). Thirdly, the inner-outer structuring is not a one-way street, for the outer can also be structured by the inner (as in the case of invention and certain acts of spontaneous play). In other words, the dependence of the inner on the outer is, though massive, not slavish. (Proper treatment of these three issues would go far beyond the scope of this essay.)
In doing so, I do not mean to bemoan the ‘Fall of Thought’. The learning of a language or acquisition of an intellect must be an essentially positive and enabling development on any account, since any account can itself only be given in language. Calling the essentials of language into doubt would be self-defeating.
The foregoing four paragraphs are a partial summary of a long discussion in my “A Meeting of the Conceptual and the Natural: Wittgenstein on Learning a Sensation-Language”, which focuses on a special kind of “naturalism” in Wittgenstein. A particular point elaborated in this paper is that the aforementioned instinct to use language is not primitive instinct but tutored instinct and, similarly, the aforementioned desire to know is not primitive-natural but second-natural (in the sense of McDowell’s “naturalism of second nature” in Mind and World, especially Lecture IV). This paper also discusses the role played in the child’s learning of a sensation-language by the adults’ use of language other than questioning.
More remarks, from both the PI and elsewhere, can be adduced to show a spirit of critique toward language in Wittgenstein. In particular, this spirit pervades the Tractatus, becoming most explicit in the aphorism “All philosophy is ‘critique of language”’ in 4.0031. For a patient interpretation of this aphorism, see Michael Kremer’s “Russell’s Merit”.
This attitude is well illustrated by a scene reported by Rush Rhees (Philosophical Occasions, p. 326). Wittgenstein once began a lecture by quoting, from a detective story magazine, a metaphysical remark about time: “A clock is a bewildering instrument at best: measuring a fragment of infinity; measuring something which does not exist perhaps.” Rhees reports: “[Wittgenstein] said it is much more revealing and important when you find this sort of confusion in something that’s said ‘in a silly detective story than it is when you find it in something that’s said by a silly philosopher’.” Wittgenstein, we might say, is especially alert to the voices of our natural disposition toward metaphysics (what Kant called metaphysica naturalis).
References
Anscombe, G. E. M. (1981). Collected philosophical papers. Oxford: Basil Blackwell.
Anscombe, G. E. M. (1957). Intention. Oxford: Basil Blackwell.
Bar-On, D. (2004). Speaking my mind: Expression and self-knowledge. Oxford: Oxford University Press.
Canfield, J. (2001). Private language: The diary case. Australasian Journal of Philosophy, 79(3), 377–394.
Cole, J. (1997). On ‘Being Faceless’: Selfhood and facial embodiment. Journal of Consciousness Studies, 4(5/6), 467–484.
Diamond, C. (2000). Does Bismarck have a beetle in his box? The private language argument in the Tractatus. In A. Crary & R. Read (Eds.), The New Wittgenstein (pp. 262–292). London: Routledge.
Engstrom, S. (2006). Understanding and sensibility. Inquiry, 49(1), 2–25.
Finkelstein, D. (2003). Expression and the Inner. Cambridge, MA: Harvard University Press.
Fogelin, R. (1996). Wittgenstein’s critique of philosophy. In H. D. Sluga & D. G. Stern (Eds.), Cambridge companion to Wittgenstein (pp. 34–58). Cambridge: Cambridge University Press.
Hacker, P. (1990). Wittgenstein, meaning and mind: An analytical commentary on the Philosophical Investigations (Vol. 3). Oxford: Basil Blackwell.
Kremer, M. (2012). Russell’s merit. In J. Zalabardo (Ed.), Wittgenstein’s early philosophy (pp. 195–240). Oxford: Oxford University Press.
McDowell, J. (1998). One strand in the private language argument, first published in 1989, later collected in his Mind, Value, and Reality. Cambridge: Harvard University Press.
McDowell, J. (1994). Mind and world. Cambridge, MA: Harvard University Press.
McDowell, J. (2009). Avoiding the myth of the given. Having the world in view: Essays on Kant, Hegel, and Sellars (pp. 256–272). Cambridge, MA: Harvard University Press.
Sellars, W. (1963). Empiricism and the philosophy of mind, first published in 1956, collected in his Science, Perception and reality (pp. 127–196). Atascadero, CA: Ridgeview Publishing Company.
Strawson, P. F. (1959). Individuals. London: Routledge.
Tang, H. (2014). ‘It is not a something, but not a nothing either!’—McDowell on Wittgenstein. Synthese, 191(3), 557–567.
Tang, H. (forthcoming). A meeting of the conceptual and the natural: Wittgenstein on learning a sensation-language. Philosophy and Phenomenological Research.
Williams, M. (1999). Wittgenstein, mind and meaning: Toward a social conception of mind. London: Routledge.
Wittgenstein, L. (1980). Culture and value (G. H. von Wright, Ed., P. Winch, Trans.). Chicago: University of Chicago Press.
Wittgenstein, L. (1982). Last writings on the philosophy of psychology (Vol. 1, G. H. von Wright & H. Nyman, Eds., C. G. Luckhardt & M. A. E. Aue, Trans.). Oxford: Basil Blackwell.
Wittgenstein, L. (1967). Philosophical investigations (G. E. M. Anscombe, Trans.). Oxford: Blackwell.
Wittgenstein, L. (2009). Philosophical investigations, revised (4th ed., P. Hacker & J. Schulte, Eds.). Oxford: Blackwell.
Wittgenstein, L. (1993). Philosophical occasions (J. C. Klagge & A. Nordmann (Eds.)). Indianapolis: Hackett Publishing Company.
Wittgenstein, L. (1974). Tractatus logico-philosophicus (D. Pears & B. McGuinness, Trans.). Paris: Routledge & Kegan Paul.
Wittgenstein, L. (1967). Zettel (G. E. M. Anscombe & G. H. von Wright, Eds., G. E. M. Anscombe, Trans.), Oxford: Basil Blackwell.
Acknowledgments
I wish to thank John McDowell for comments on an earlier draft, and two anonymous referees for Synthese for very helpful comments. I also wish to thank the Despoina-Stiftung for special support, and the Fundamental Research Funds for the Central Universities (a Chinese funding agency) for financial support.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Tang, H. Wittgenstein and the Dualism of the Inner and the Outer. Synthese 191, 3173–3194 (2014). https://doi.org/10.1007/s11229-014-0441-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-014-0441-2