Antonio Damasio looks forward to a ‘cognitive neuroscience of belief’ in his concluding remarks to this admirable collection, because ‘understanding the nature of belief is essential for making sense of human behavior in a comprehensive manner’ (p. 325). In taking a few first steps toward this goal, Memory, Brain and Belief could inaugurate a new field, one that resembles ‘consciousness studies’ in both its interdisciplinary nature and its problems of self-definition. Editors Schacter and Scarry admit that ‘we are still some distance from an adequate working definition of belief that is shared across scientists and scholars’ (p. 4), and the study of belief as presented here looks more embryonic than comprehensive. But the high standards of both science and scholarship offered all the way from introduction to index make this book an auspicious beginning.
If a new field of belief has been born here, it is still umbilically connected to the mother field of memory, and more specifically to the high-profile issue of ‘false memory.’ Schacter's Searching for Memory (1996) was in part a humanely scientific response to this public concern; meanwhile both Schacter and Scarry took part in a Harvard-based working group that ‘focused intensively on the nature of memory distortion’ (p. 2). This group organized the 1997 conference which culminated three years later in Memory, Brain and Belief. The nature of belief became an issue in the context of false memories because, as the editors remark, ‘perhaps the most striking feature of such inaccurate memories is that they are often expressed as powerful, seemingly unshakable beliefs about the past’ (p. 2). Since these memories are both autobiographical (or ‘episodic’) and problematic, the book retains a double focus on autobiography and pathology, even while it points the way toward a wider concern with ‘belief’.
Autobiography as a literary form reflects the language-mediated process by which we construct episodic memories of our own past—and thus ourselves as continuing beings. The ‘constructivist’ consensus shared by all contributors here is articulated by Paul John Eakin in his reference to autobiography as ‘fiction’, using the term not in a pejorative sense but ‘in its root meanings: that which is formed, shaped, molded, fashioned, invented’ (p. 290). The self is a fiction in the same sense, because it is constructed through narrative and because ‘memory, the would-be anchor of selves and lives, constructs the materials from the past that an earlier, more innocent view would have us believe it merely stored’ (p. 291). In a similar vein, Sissela Bok speaks of autobiography as a ‘moral battleground’, with particular reference to the works of Virginia Woolf and the Tolstoy family. Michael Ross and Anne Wilson also explore the construction and appraisal of ‘past selves’, this time from a sociological perspective.
Endel Tulving and Martin Lepage, reviewing research on the ‘special neural mechanisms’ which subserve episodic memory, celebrate it and the ‘autonoetic consciousness’ which emerges from it as ‘true marvels of nature’ (p. 225) because they enable us to ‘travel back and forth in subjectively experienced time’ (p. 216). Katherine Nelson explains how this ability develops in tandem with language, ‘theory of mind’ and self-concept during the first five years of a child's life, and her account (unlike most of those mentioned so far) highlights the role of belief. She shows that the attribution of belief states is inextricable from the whole development process, and one of the essential tools for investigating this process is the ‘false belief test’: if a child does not recognize that others can have false ideas about a situation, there is no reason to think that he has any conception of conception, perception, or thought in others or himself.
Nelson also points out that preschool children are especially prone to ‘source errors in memory’ (p. 275). This relates the developmental story to the ‘false memory’ problem, via the ‘source monitoring framework’ which is explained by Marcia Johnson and Carol Raye in the book's most exhaustively researched article. Johnson and Raye account for ‘false memories and beliefs’ as various kinds of breakdown in normal evaluation processes. We assess the content of our own memories and beliefs by trying to track where they come from (actual past experience, stories we tell ourselves or others, stories they tell us, and so on). But source monitoring is complex, recursive, difficult, time-consuming, easily disrupted or misdirected, and often inconclusive even for a healthy and mature mind/brain. Thus it is not surprising that ‘source errors’ are involved in many of the pathologies explored in this book.
Pathology often provides crucial clues in the investigation of normal processes; where would the study of memory be without case studies relating specific brain lesions to specific forms of amnesia? So it seems reasonable to expect that pathologies of belief will open windows upon its normal workings. To this end, V.S. Ramachandran cites some extreme cases of false belief—such as the anosognosic stroke victim who insisted that she was pointing at his nose with her left arm even as it lay paralyzed in her lap—and offers his own theory about the brain mechanisms involved. Chris Frith and Raymond J. Dolan explore the role of memory in the delusions associated with schizophrenia, while Mahzarin Banaji and R. Bhaskar look into the social causes and effects of implicit stereotyping. Here we are looking mainly at beliefs about the world (i.e. ‘semantic’ memory) rather than beliefs about the ‘veridicality’ of episodic memories.
The problem with this approach is the temptation to define ‘belief’ as a pathology, a trap which seems to ensnare Howard Eichenbaum and J. Alexander Bodkin. They show that damage to the hippocampus in rats prevents them from using visible landmarks as normal rats do when learning to navigate a water maze, and they explain this with a salient contrast between two kinds of memory. But instead of using available terms (perhaps ‘semantic’ and ‘procedural’) to name the contrast, they dub the more adaptive kind of memory ‘knowledge’ and define ‘belief’ as ‘a disposition to behave in a manner that is resistant to correction by experience’ (p. 177). Despite their attempt at a rationale, even those who don't mind attributing beliefs to rats may find this choice of terms arbitrary. In ordinary discourse, people typically say that they ‘believe’ something in order to express diffidence about it; if they want to express inflexible attachment to an idea they are more likely to claim that they ‘know’ it. Eichenbaum and Bodkin reverse this usage, thus undermining their own contribution.
The one article here that steps completely outside the frames of autobiography and pathology is the opening one by Chris Westbury and Daniel Dennett. Aiming to respect established usage while avoiding ‘philosophically problematic’ assumptions, they define belief not ‘as an entity that an organism might have or not have’ but ‘in terms of the circumstance under which a belief can be justifiably attributed to that organism’ (p. 24). This is compatible with Nelson's developmental account and with Damasio's more rough-and-ready definition of belief as ‘attribution of truth value to a particular thought content’ (p. 326), while leaving open the partly empirical question of what motivates us to make such attributions.
A science of belief will have to raise some questions that are not addressed in Memory, Brain and Belief. For instance, not even Eakin's discussion of ‘fiction’ deals with the peculiar response to literature that Coleridge called the ‘willing suspension of disbelief.’ The vast regions of myth and religion also go unexplored in this book, a curious omission given that ‘religious studies’ was among the disciplines represented in the working group which gave rise to it (p. 1). But a new discipline, like the proverbial thousand-mile journey, has to begin with the first step, and this one is very well taken.
Gary Fuhrman