Simon Prosser and Francois Recanati, eds.: Immunity to Error through Misidentification.
Immunity to Error through Misidentification.
Cambridge: Cambridge University Press 2012.
$95.00 (cloth ISBN 978-0-521-19830-1)
When essaying a singular judgment of the form a is F there seem to be two distinct ways of falling into error, in principle at least. First, one may be wrong about whether the object a really is F. Second, one may be wrong about whether it is a that is F; one reaches for the right predicate, but misidentifies the object in question when applying it. In a well-known passage in The Blue Book, Wittgenstein pointed out that certain judgments seem to have the property of being immune to the second kind of error distinguished above: the property of immunity to error through misidentification (IEM), to adopt now standard terminology introduced by Sydney Shoemaker. This excellent volume offers thirteen new essays on IEM, which collectively attempt to get clearer on the nature and scope of the phenomenon and to discuss its bearing on questions about de se thought and indexical thought more generally, as well as self-knowledge, the metaphysics of selves, and other related issues in epistemology and the philosophies of mind and language.
John Campbell tries to identify an insight in the thesis, primarily associated with the later Wittgenstein and with Elizabeth Anscombe, that 'I' is not a referring term. This insight, Campbell argues, is that the way the referent of 'I' is determined does not causally or normatively explain the 'pattern of use' we make of it. In particular, and in contrast with descriptive names (such as Evans's 'Julius') and perceptual demonstratives, the way the referent of 'I' is determined doesn't explain why some judgments involving the first-person are IEM, while others aren't. I wasn't quite clear on what Campbell meant by a 'pattern of use'. He offers the example of the introduction and elimination rules for the logical constants, which are normatively explained by the truth tables. But I couldn't quite see in what sense IEM might be part of the pattern of use of 'I', taking this as the model.
Annalisa Coliva considers the source of philosopher's tendency towards semantic, metaphysical and epistemological extravagance when theorizing about the first-person. She makes three important points with which I'm very much inclined to agree. First, she argues that Peacocke's dismissal of the suggestion that IEM is a component is too quick and his own diagnosis solely in terms of 'representational independence' of certain I-judgments--that they involve self-applying the first-person concept but are not grounded in first-person contents--is inadequate. Second, she argues that in order to understand how IEM is implicated in the tendency towards extravagance, one must appeal to the distinction gestured at with the (misleading) labels 'logical' and 'de facto' IEM. Third, Coliva suggests that recognizing that some psychological self-attributions are logically IEM and explaining this feature need not generate any real pressure to abandon the commonsense view that 'I' refers to human being.
The contribution by Marina Folescu and Jim Higginbotham is both the shortest paper in the volume and the most technical and difficult. They start by questioning both the force of the standard familiar examples designed to show that the de se is a distinct phenomenon and the adequacy of David Lewis's well-known treatment of it in terms of de se thoughts involving the self-attribution of properties. They then appeal to semantic considerations from both English and Romanian to argue that there are two different sorts of de se contexts, each giving rise to IEM judgments in distinct ways. Their semantic observations are highly interesting but the paper ends very abruptly, leaving more philosophical questions than answers. In particular, the import of any of this for discussions of IEM, or indeed, exactly what Folescu and Higginbotham take IEM to be, is left rather unclear.
Jenann Ismael appeals to information theory to try to vindicate the validity of the two transitions she takes to lie at the heart of the puzzle generated by IEM. The first involves the idea that IEM judgments are based on 'thetic'--subjectless--contents: this is related to Peacocke's notion of representational independence and a suggestion of Recanati's discussed below. The other transition is from a self-ascription to a third-person ascription, such as 'AM has a headache'. The puzzle is that a subjectless content can ground a self-ascription which is IEM, but which is made using the first-person pronoun, which 'exhibits all the characteristics of a genuinely referring term'; '[i]t occupies an argument place that can take other values, it is intersubstitutable with other representationally mediated forms of reference to oneself, and it supports all of the ordinary argument patterns' (63). It's not really clear what cash-value the appeal to information theory has, since most of the work seems to be done by the suggestion that what each of us implicitly means by 'I' is 'the person to whose psychological history these very thoughts and experiences belong' (75).
Wittgenstein introduced IEM as a distinctive feature of the use of 'I' as subject rather than object, and one might wonder whether this is just Kant's distinction between consciousness of oneself as subject and of oneself as an object, and if it is not, how Kant's distinction bears on questions about the nature and source of IEM. This is precisely the topic of Beatrice Longuenesse's contribution. She argues convincingly that the distinctions do not quite line up, since many of Wittgenstein's uses of 'I' as subject would count as expressions of consciousness of oneself as an object by Kant's lights. She takes this to suggest that there are two distinct uses of 'I' as subject, each of which gives rise to IEM judgments in a different way. The paper closes with a discussion of Oliver Sacks's intriguing case of the 'disembodied woman'--a woman who stopped gaining any information about the position of her body from proprioception--and the relationship between Kant's consciousness of oneself as subject and Freud's 'ego'.
Daniel Morgan defends what he calls the Simple Explanation of IEM, according to which IEM judgments are just those that are not based on an identification, and argues on this basis that it sheds no light on de se thought. He takes as stalking horses Higginbotham's reflexive account of IEM and Recanati's appeal to implicit de se thoughts. Morgan argues that these attempted explanations of IEM trip up in ways that the Simple Explanation does not, and that in any case these views fail to gain motivation from consideration of IEM, since the Simple Explanation does the trick without needing to appeal to an exotic account of the nature of de se thought.
In her contribution Lucy O'Brien revisits her earlier argument for the twin conclusions that (a) self-ascriptions of action possess a stronger kind of IEM than self-ascriptions of bodily movement do and that (b) providing the basis for judgments possessing this stronger kind of IEM is required for a way of knowing about oneself to be source of self-knowledge. O'Brien now finds her earlier argument dubious, and her criterion for self-knowledge too demanding. She proposes that what crucial for a source of self-knowledge is not that that a subject can always discriminate it from one that gives information about another person, as she used to think, but rather it delivers information that is necessarily first-person in content (132). O'Brien closes with an interesting but much too brief discussion of how we might argue directly for the claim that action awareness is essentially first-personal in a way that bodily awareness is not.
In order to explain why first-person judgments display IEM in normal circumstances, Christopher Peacocke appeals to his view that concepts are individuated by their fundamental reference rules together with a proposal for the rule for the first-person concept. (His main thesis therefore stands in direct opposition to Campbell's.) According to Peacocke, the fundamental rule for the first-person concept is 'What makes someone the reference of I in a thinking is that he or she is the producer of that thinking' (145), and Peacocke argues that appealing to this rule enables one to explain how a subject can reasonably self-ascribe certain properties in ways that do not rest on an identification, or which rest only on identities which are a priori and constitutive or which characteristically hold in normal conditions (148-9).
In his piece, Simon Prosser points out that there are errors involving the subject of a judgment rather than the predicate that are not errors through misidentification. For example, if I judge that that woman is walking slowly on the basis of a close perceptual encounter, my judgment is plausibly IEM; while I might be wrong because the person I see is in fact walking quite quickly, there doesn't seem to be any sense to be made of the idea that I've misidentified the person walking slowly as that woman. But of course, there still is scope for me to have made some kind of mistake, even if my judgment is IEM and my predication is entirely in order, since the person I've seen may be male rather than female. (Wright makes a similar observation at 247-8.) Prosser calls this an error through misclassification, and he argues that judgments involving most indexicals and demonstratives are either IEM or immune to error through misclassification, but not both. Perhaps the main interest in this is the new light it promises to shed on the first-person when one hypotheses that I functions much like the other indexicals Prosser discusses, and the paper closes with a brief discussion of how the resulting view compares to what others have held about I.
In previous work Francois Recanati has distinguished primitive IEM from derivative IEM. IEM judgments that are 'directly based on experience' (183) will be primitively IEM, but judgments that are not directly based on experience can inherit immunity if based on another IEM judgment. De se judgments will be primitively IEM when directly based on experiences that are 'intrinsically first-personal' (185). Such experiences are thetic, in the sense introduced above; they are first-personal not in virtue of representing the experiencing subject, but because the 'mode' of the experience--'the manner in which a content is present to or entertained by the mind', as Wright explains the idea (262, n. 18)--ensures that it must concern the subject of that experience. On this basis, Recanati had proposed that only implicit de se thoughts--thoughts that are de se because their content is evaluated with respect to the subject rather than because that subject is explicitly represented in their content--can be IEM. He now takes this to be clearly false, and the aim of his paper here is to extend his account to explicit de se thoughts. The central idea is that no additional evidence is needed for a subject to judge the content that makes explicit what was already implicit in the mode of the experience grounding that judgment. He then turns to generalizing the account to demonstrative thoughts and indexical thoughts more generally.
In Crispin Wright's response (which falls 50 pages after Recanati's paper due to an unfortunate decision to order the papers alphabetically by author) Wright wonders how exactly Recanati secures the result that a de se judgment made on the basis of an experience with an essentially first-personal mode will be IEM (264-7). He argues that Recanati's account clearly fails to generalize to indexical thoughts which are IEM but which are not plausibly thought of as grounded in experiences that carry thetic content (273-7). Instead, Wright advocates what he calls the Simple Account of IEM, which is much like the Simple Explanation defended by Morgan.
Unfortunately, Galen Strawson seems to work with a different conception of IEM from the standard one that figures in the other papers in this volume. In fact, it becomes plain early on in his paper that by IEM he really means what others call guaranteed referential success (203-4). In effect, he tries to reconcile this guaranteed success with his suggestion that one's uses of I sometimes refer to a complete human being and sometimes to a self, which he suggests is 'a pattern or 'synergy' of neural activity, possibly short-lived' (209). I found Strawson's discussion of why I cannot be univocal rather quick and unpersuasive, and so I wasn't persuaded that Strawson's project in this paper was well-motivated, or that it really concerned the topic of the volume as directly as it might have.
Frederique de Vignemont carefully considers recent empirical work suggesting that bodily self-ascriptions are not IEM, including the Rubber Hand Illusion and variants, cases of people suffering from somatoparaphrenia (the delusion that a part of one's body--typically a limb--does not belong to one), and more generally evidence showing that our knowledge of our own bodies is multimodal. She argues that we can preserve claims of bodily IEM in the face of these empirical considerations, so long as we are willing to pay close attention to the kinds of grounds that give rise to IEM judgments about one's own body. The discussion here is fascinating, and many of de Vignemont's points are convincing, though her treatment of the somatoparaphrenia cases struck me as rather too quick.
Sadly, this volume won't make for a good introduction to the topics it discusses. The editors do provide an introduction, but it is very brief and doesn't attempt to provide any kind of overview of the issues. As I've already hinted, I also found the decision to organize the collection by the surnames of the authors rather than by theme unfortunate. And it has to be said that the book is not a remotely easy read, with most of the papers demanding to be read two or three times before one has a clear sense of what's going on. But the papers do reward that kind of careful attention, and the volume as a whole stands as an important contribution to scholarship on immunity to error through misidentification and neighbouring philosophical questions. It will no doubt be a major source of ideas and inspiration for future work on these issues.
University of Edinburgh
|Printer friendly Cite/link Email Feedback|
|Publication:||Philosophy in Review|
|Article Type:||Book review|
|Date:||Aug 1, 2013|
|Previous Article:||Olga Pombo, Juan Manuel Torres, John Symons, and Shahid Rahman, eds.: Special Sciences and the Unity of Science.|
|Next Article:||Theodore Sider: Writing the Book of the World.|