Kevin MacNeil, Culver Academies

Transcription

1 112 Philosophical Investigations beyond the boundaries of what we can know, but which must be understood in personal terms (142), then is not this already all that is really needed for a reasonable and well-grounded religious faith? However that may be, there are ample reasons to welcome this beautifully written and erudite essay, which provides a superbly clear conspectus of Roger Scruton s current philosophical and religious thinking. University of Reading and Heythrop College London jgcottingham@mac.com Michael Hymers, Wittgenstein on Sensation and Perception (Routledge, 2017). xiii + 202, price hb. Kevin MacNeil, Culver Academies On the standard reading, Wittgenstein began to abandon the picture-theory through conversations with Frank Ramsey and others in the course of which he came to see insuperable difficulties with the idea of propositional independence, central to the Tractarian account. But one has only to look to Some Remarks on Logical Form or the famous Christmas Day letter to Friedrich Waismann to see that Wittgenstein was already formulating solutions to these problems, which had they been followed through to completion could have preserved the picture-theory of the Tractatus by relaxing the conditions on the logical independence of elementary propositions. It is striking that Wittgenstein foregrounds this disavowal by introducing the concept of grammatical analysis at the very beginning of Philosophical Remarks. A proposition is completely logically analysed, we are told, if its grammar is made completely clear. And, this no longer requires a unique phenomenological language whose form pictures the world. I do not now have phenomenological language, or primary language as I used to call it, in mind as my goal, observes Wittgenstein, for I no longer hold it to be necessary. Instead, phenomenology would be the grammar of the description of those facts on which physics builds its theories. (1980: 51. Italics mine.) No longer necessary? A primary language of the sort Wittgenstein envisioned in the Tractatus is no longer thought possible. The focus has shifted, instead, to our language, and it should be clear that elucidating this language thereby separating what is essential from what is inessential cannot be done by means of the picture-theory. The work once done by the picture-theory is now to be accomplished, instead, by a philosophical grammar.

2 Kevin M. MacNeil 113 By way of illustration, Wittgenstein introduces the colour octahedron of the Austrian psychologist Alois H ofler. With the pure colours at the corner points, H ofler s double tetrahedron provides a rough representation of colour-space. Significantly, this is a grammatical representation, not a psychological one (1980: 51), and it is a matter of grammar rather than experience that we can speak of a greenish blue but not a greenish red (1982: 8). THE COLOUR OCTAHEDRON (Wittgenstein 1982: 8) A rough representation, indeed! So, we can talk about greenish blue or bluish green, and even green-blue, if we have in mind an imprecise interval of colour. But, what if our purposes demand greater accuracy? Wittgenstein presents a more refined model. THE DOUBLE EIGHT-SIDED PYRAMID (Wittgenstein 1980: 278) We can now speak meaningfully both of primary and secondary colours. Wittgenstein is quick to point out that the propositions the primary colour

3 114 Philosophical Investigations red is between the secondary colours orange and magenta and the secondary colour magenta is between the primary colours red and blue capture different senses of the word between. But, this is already reflected in the notation. So which system is the correct one? Neither must and either may be the model of colour-space. No single picture compels assent, as the author of the Tractatus once believed. No ideal language here. No privileged logic. Each representation and many others are possible carries with it its own grammar. H ofler s natural colour system was designed to model colour perception. What if we are interested in the physics rather than the physiology of colour? Then, the focus shifts to triadic primary structures redgreen-blue for the additive mixing of light, and yellow-magenta-cyan for the subtractive mixing of pigment with the primary elements of one serving as the secondary components of the other. Are red and yellow, then, primary colours? What is our target? Perception? Light? Pigment? Their grammar depends upon our purposes. What led Wittgenstein to abandon the picture-theory if not difficulties with elementary propositions? His remarks on the grammar of colour concepts invite a more fruitful line of inquiry. So, the issues raised by Michael Hymers in Wittgenstein on Sensation and Perception are of the first importance. Munro Professor of Metaphysics at Dalhousie University and past editor of the Canadian Journal of Philosophy, Hymers is the author of Philosophy and Its Epistemic Neuroses (2000) and Wittgenstein and the Practice of Philosophy (2010). A careful study of Wittgenstein s philosophical method, Hymers second book is considered by David Stern of the University of Iowa to present a wide-ranging and extraordinarily clear introduction to Wittgenstein s treatment of central philosophical problems. In the work currently under discussion, Michael Hymers advances two theses one interpretive, the other philosophical grounded in Wittgenstein s reflections on sense-datum theories and the misleading metaphor of phenomenal space. In such works as Philosophical Remarks and The Big Typescript, from the early transitional period following Wittgenstein s selfimposed philosophical exile, Hymers traces a rapidly evolving argument, from the phenomenology of sense-data to the grammar of appearance, that forms a powerful critique of the metaphor in question. Hymers then argues for Wittgenstein s relevance to contemporary debates in the philosophy of perception, including sense-datum theories, of course, but also recent work on sensory qualia and the authority of first-person accounts of immediate experience. Along the way, Hymers presents detailed discussions of first-person sensation reports, intentional attitudes, phantom pain, univocality and indeterminacy, Frank

4 Kevin M. MacNeil 115 Jackson and Howard Robinson on sense-data, Sydney Shoemaker and Ned Block on inverted spectra. In what follows, we will focus on Hymers interpretive claim, for it forms the foundation of his philosophical argument and is likely, in any case, to be of greater interest to Wittgenstein scholars. What is the relationship between the private space of sensation and the public space of physical objects? As Hymers has it, Wittgenstein pursues three successive strategies to address this question, a novel contention, for sure, and one that stands at at the [very] heart of Hymers reading of the central texts of Wittgenstein s early interlude (2017: 2). A propos of the first strategy, Hymers makes much of the following entry in the Nachlass, believing it to present evidence that Wittgenstein, as of October 1929, considered the phenomenological language project doomed : The assumption that a phenomenological language is possible and that only it would really say what we want to express in philosophy is I believe absurd. We must make with our ordinary language and only understand it aright, i.e. we must not let it mislead us into thinking nonsense. (2017: 43) But these remarks do not signal the rejection of phenomenological language tout court, merely the repudiation of an ideal language of the Tractarian type. As Wittgenstein has it: The phenomenological language describes precisely the same thing as ordinary, physical language. Only it must restrict itself to what is verifiable (2017: 41). We cannot have a greenish-red patch. We cannot have a greenish-red chair. Hymers views phenomenal language as private language, and is therefore perplexed by what he sees as Wittgenstein s attempt in Some Remarks on Logical Form to mathematize immediate experience. I doubt there is any coherent way of fitting all these pieces together, Hymers concedes (2017: 42). We have many reasons to reject the alleged mathematization project, but not on grounds of privacy. A phenomenal language, as we have seen, will share certain grammatical features with ordinary, physical language red patch, red chair so there can be no question of its privacy. Throughout the period in question, Wittgenstein consistently held the view that the word sense-datum really means the same as appearance. But he acknowledged that the term could be misleading. If personification means, e.g. using the word time as though it were the name of a person, then objectification is talking of it as though it were a thing. No puzzles arise out of personification, Wittgenstein

5 116 Philosophical Investigations claims. No one actually thinks that the ship is female. Whatever one makes of this, Wittgenstein clearly believed that things fell otherwise with objectification, for here there is a tendency to assimilate the grammar of appearance to the grammar of physical objects (1993: 312). And we find ourselves thinking if something seems to be red then something must have been red; if something seemed to last for a short time, then something must have lasted for short time; etc. (2013: 348e). This, of course, is a grammatical mistake. Hymers worries about the frequent, uncritical mentions of sensedata in student lecture notes from the early s, conjecturing that Wittgenstein adopted a form of conventionalism about sense-data in response to the pervasiveness of sense-data talk in the Cambridge of the day (2017: 94). But this worry seems misplaced. For the term sensedata now functions very much as the term object once did, as an element in the grammar of the description of the language on which physics builds its theories (1980: 51). This language includes propositions like This is red or That is a brown patch. But not Sense-data are the source of our concepts (1982: 81). That is a grammatical remark. So, there is no second strategy, according to which we can talk about sense-data, so long as we remember that everything we say must be translatable into ontologically non-committal claims about how things seem or appear (2017: 94). In fact, we do not talk about sense-data at all. We talk about high pitches or red patches, and use the language of sense-data to clarify what can be said. Hymers is surely right to contend when sense-data are seen aright that we have little reason to search for a private language of sensation and perception that represents a phenomenal space of private objects. And, one can hardly claim privileged access to a space one no longer believes in. Avowals of pain and other first-person ascriptions must be explained otherwise. But, we have argued that Hymers so-called third stage was actually operative in Wittgenstein s thinking from the late s onwards. So, it becomes difficult to endorse the idea that we miss the point of the so-called private-language argument unless we see that it is rooted in Wittgenstein s critique of the misleading metaphor of phenomenal space (2017: 94 95). We are on safer ground to say that both the critique of the metaphor and the private-language argument emerge out of Wittgenstein s increasingly sophisticated post-tractarian investigations into the grammar ordinary language. That said, Michael Hymers provides a sustained and valuable analysis of the metaphor of phenomenal space, arguing convincingly that Wittgenstein, from the early s onwards, understood the pitfalls of taking this trope literally. One of the many virtues of Hymers work is that it draws attention to a highly productive but little understood period

6 of Wittgenstein s philosophical development, one that merits greater scholarly attention. Hymers meticulous study will be of interest to anyone seeking a better understanding of Wittgenstein s early work on the grammar of sensation and perception. Culver Academies 1300 Academy Road #155 Culver, IN USA kevin.macneil@culver.org Kevin M. MacNeil 117 References Wittgenstein, L. (1980). Philosophical Remarks. R. Rhees (eds.), R. Hargreaves and R. White, trans., Chicago: University of Chicago Press.. (1982). Wittgenstein s Lectures, Cambridge, : From the notes of John King and Desmond Lee. Chicago: University of Chicago Press.. (1993). The Language of Sense Data and Private Experience (notes by Rush Rhees) Philosophical Occasions: J. Klagge and A. Nordmann (eds.), Indianapolis: Hackett Publishing Company.. (2013). The Big Typescript: TS 213. C. Grant Luckhardt and A. E. Maximilian (eds.), Aue. Chichester, West Sussex, UK: Wiley- Blackwell.