The Repeated Reproduction of Bartlett’s Remembering

Elizabeth B. Johnston

Sarah Lawrence College

Abstract

There is a striking parallel between the treatment of Bartlett's theories of memory in the psychological literature and Bartlett's own characterization of reproductive memory as interest driven and constructive. Three periods of intensified interest in Bartlett’s classic book Remembering can be identified. The first occurred in the wake of the publication of Remembering and focused on replication and extension of the empirical work. The second was during the period of the 'cognitive revolution' and treated Bartlett's key theoretical concept of 'schema' within an information processing framework. The third is an ongoing revitalization of interest in the cultural and social aspects of Bartlett's multifaceted theory. For each era I select one influential study that provides a metonymic and transformative reproduction of Remembering to illustrate that the treatment of Bartlett's work at different times provides a barometer of prevailing opinion in experimental psychology.

Introduction

Interest in Bartlett’s book Remembering has been lively since its publication in 1932. Psychologists have revisited Bartlett’s work at regular intervals, and reproduced the experiments with variations and elaborations (Northway, 1936; Oldfield & Zangwill, 1943; Allport &Postman, 1947; Kay, 1955; Gomulicki, 1956; Paul, 1959; Johnson, 1962; Gauld & Stephenson, 1967; Mandler & Johnson, 1977; Edwards & Middleton, 1988; Wheeler & Roediger, 1992; Roediger, Wheeler & Rajaram, 1993; Saito, 1996; Wynn & Logie, 1998). The book has been reissued twice, in 1964 and 1995, a notable fact in this era of short print lifespans. It is consistently referred to as a classic in the one to two page descriptions provided in the majority of introductory and cognitive psychology texts, from the time of its publication to the present day.

Bartlett’s experimental technique was to ask his subjects to repeatedly reproduce drawings, stories or prose passages. His most famous material is a North American folktale entitled The War of the Ghosts that he appropriated from Boas (Beals, 1998; Boas, 1901; see Appendix for Bartlett’s version of the story). Bartlett used two techniques: the method of repeated reproduction where his subjects were asked to provide a number of renditions of the tale, at long intervals, as much as ten years in one case; and the method of serial reproduction, fashioned after the parlor game of ‘Russian Scandal’ or ‘Telephone’, as it was later known, where subjects passed along the story in a chain.

Bartlett’s famous reproduction technique has been reenacted in the psychological literature with Bartlett’s work as the text. The purpose of this paper is to set out the ways in which the principles of remembering outlined by Bartlett provide appropriate descriptions of the subsequent treatment of his work. This necessitates another revisiting of Bartlett to provide my version of his findings.

Bartlett’s Findings in Historical Context

[I]f I am to say what sort of psychologist I am, I think I can say only that I am a Cambridge psychologist. The trouble about this is that Cambridge psychology of the laboratory type has never committed itself to any hard and fast and settled scheme of psychological explanation. I hope it never will. (Bartlett, 1936, p. 40)

It is clear from both Crampton's (1978) doctoral thesis, The Cambridge School: The life, work and influence of James Ward, W.H.R. Rivers, C.S. Myers and Sir Frederic Bartlett, and Bartlett's writings on Cambridge psychology (Bartlett, 1936, 1937, 1955) that a group of four fellow Cambridge men profoundly influenced the development of Bartlett's thought: James Ward, W.H.R. Rivers, C.S. Myers and Henry Head. James Ward (1843-1925) was Professor of Mental Philosophy and Logic at Cambridge when Bartlett came as an undergraduate student. Ward studied with Lötze in Göttingen; Lötze's Medicinishe Psychologie, oder Physiologie der Seele (Medical Psychology, or The Physiology of the Soul) was one of the early texts that contributed to forming the discipline of psychology through an integration of philosophy and physiology. Ward brought the work of the new German experimental tradition to Britain by translating the work of Weber and Fechner on psychophysics. Bartlett describes Ward's lectures as "Germans, Germans all the way." (Bartlett's unpublished autobiography cited in Crampton, 1978). Prior to his undergraduate work at Cambridge Bartlett reports cycling 18 miles a week to read Ward's famous article on 'Psychology' in Encyclopedia Britannica in installments. In this widely read article Ward attacked atomistic associationism and developed a new psychology that emphasized the role of the active organizing subject. Northway (1940) describes Bartlett's psychology as essentially Wardian, and Ward's influence is plain in several aspects of the view of memory Bartlett put forward in Remembering.

First, memory is an active, dynamic, inferential process that is better characterized as constructive than reproductive. The first point Bartlett makes in his summary of the repeated reproduction results is: "accuracy of reproduction, in a literal sense, is the rare exception and not the rule." (Remembering, p. 93) This constructive process is memorably described by Bartlett thus:

It is fitting to speak of every human cognitive reaction - perceiving, imaging, remembering, thinking and reasoning - as an effort after meaning. (Remembering, p. 44)

The constructive nature of recall leads to the following changes in any reproduction of material: omissions, condensations, elaborations, transpositions, transformations, and, less frequently, importations. As a counterpoint to his focus on invention and importation, Bartlett recognizes the strong countervailing tendency of his subjects to stereotypy. In summarizing the repeated reproduction experiments Bartlett states:

The most general characteristic of the whole of this group of experiments was the persistence, for any given subject, of the ‘form’ of his first reproduction. (Remembering, p. 83)

Ward's focus on the active subject is also apparent in the attention Bartlett paid to individual attitudes and responses. Bartlett was fascinated by the effect of ‘established interests’ on subsequent recall. The text of Remembering is packed with examples of people’s occupations directing their perceptions and memory: early on he offers the hypothetical example of the differences in what is noted by a landscape artist, a naturalist, and a geologist walking in the country (p. 4); a mathematician notes that a squiggle reminds him of a determinant and recalls it accurately several weeks later (p. 21); a minister sees Nebuchadnezzer’s fiery furnace in an inkblot (p.38); the same blot reminds a physiologist of ‘an exposure of the basal lumbar region of the digestive system as far back as the vertebral column up to the floating ribs’ (p. 38); an anthropologist who later specialized in kinship rationalizes the Ghosts of the War of the Ghosts story as a clan name (pp. 69-70); a painter visualizes the whole scene of the War of the Ghosts and draws a plan of his imagery (p.72); a Swazi cattle herdsman demonstrates remarkably accurate memory for a group of cattle purchased a year previously (p. 249-251); and a geologist turned mining engineer produces a good copy of a map of an area of the Belgian Congo he prospected more than a year earlier (p. 251-252). These examples contribute to the liveliness and interest of a text that still seems groundbreaking 65 years later.

In line with Ward's attack on atomistic associationism, Bartlett explicitly opposed a ‘trace theory’ of memory. This is a common and compelling view of the physical nature of memory that one investigator describes as "each separate experience leaves in the organism a characteristic and distinct physical record, which retains its separate identity and may, under appropriate conditions, exert a direct effect on subsequent behavior or permit conscious recall of the original experience." (Gomulicki, 1953, p. 47). Neisser (1967) coins the term "Reappearance Hypothesis" to characterize this viewpoint. Bartlett’s antipathy to this view is evident in the following frequently quoted passage from Remembering:

Remembering is not the re-excitation of innumerable fixed, lifeless and fragmentary traces. It is an imaginative reconstruction, built out of the relation of our attitude towards a whole active mass of organized past reactions or experience, and to a little outstanding detail which commonly appears in image or in language form. It is thus hardly ever really exact, even in the most rudimentary cases of rote recapitulation, and it is not at all important that it should be so. (p. 213)

The alternative that Bartlett proposes is schema, a term he used to refer to the influence of prior knowledge and predispositions in the active organization and subsequent memorability of new material. The Cambridge neurologist Henry Head (1861-1940) is explicitly acknowledged by Bartlett as the inspiration for this theoretical innovation (Bartlett, 1932, p. 198-200). Although Bartlett attributes his use of the term schema exclusively to Head, the concept can be traced back further. Crampton points out that it is suggested in the work of Kant, Lotze, the Wurzberg school and the Gestaltists (Crampton, 1978, p. 331). Bartlett's development of the schema concept provides a dynamic view of memory and is one of the most important reasons for Bartlett's continuing influence.

Bartlett’s opposition to the trace theory led to an equally strong abhorrence of the methodology that sprung from that viewpoint: Ebbinghaus’s famous studies of the laws of memorization, uncovered using series of nonsense syllables as the stimuli (Ebbinghaus, 1885/1964); Murray, 1976; Marshall & Fryer, 1978). The first chapter of Remembering contains an incisive critique of Ebbinghaus’s fallacious assumption of a simple correspondence between stimulus and response that disregards the human engagement with "effort after meaning". Referring to Ebbinghaus, Bartlett states:

His ideals were the simplification of stimuli and the isolation of response. He secured the first by using nonsense syllables as his memory material, and the second he curiously thought, followed immediately. (p. 2)

In opposition to Ebbinghaus’s approach Bartlett concentrated his efforts on meaningful stimuli that more closely approximated the types of material that people do have to remember in everyday life.

I endeavoured, in this series of experiments, to avoid as far as possible the artificiality which often hangs over laboratory experiments in psychology. I therefore discarded the use of nonsense syllables and throughout employed material a part of which, at least, might fairly be regarded as interesting and sufficiently normal for the subjects concerned not to force upon them ad hoc modes of observation and of recall. (p.47)

This is an early statement of an enduring and important controversy in the memory literature: the everyday/laboratory memory debate (Neisser, 1978; Banaji & Crowder, 1989).

While Ward can be considered the most important influence on the early development of Bartlett's theoretical approach to psychology, WH.R. Rivers (1864-1921) and C.S. Myers (1873-1946) were crucial to the formation of his interests. Both Rivers and Myers were physicians turned psychologists with training in the rigorous psychophysics of the German school. In addition to their extensive work in experimental psychology Rivers and Myers both treated psychiatric patients as their WWI service. Myers coined the term 'shell shock' to refer to the mental distress and disorder that resulted from front line experience and advocated psychotheraputic treatment. Rivers and Myers were familiar with Freud's writings and 'agreed that supportive psychotherapy and the recovery of repressed memories were the keys to cure shell shock' (Crampton, 1978, p. 204). The effect of this clinical work on Bartlett's conception of memory is not explicitly discussed at any point in Remembering, but the following quotation lends weight to the notion that it formed part of the background of Bartlett's thinking:

If the experimentalist in psychology once recognizes that he remains to a great extent a clinician, he is forced to realize that the study of any well developed psychological function is possible only in the light of a consideration of its history. (Remembering, p. 15)

The developmental - or genetic - focus in the above quotation also hints at continuities between Bartlettian and Piagetian psychology. Their common use of the key theoretical concept of 'schema' reinforces this association.

A distinct and more important influence of Rivers, and to a lesser extent Myers, on the formation of Bartlett's psychology was to stimulate his interest in anthropology. Inspired by Rivers' work on kinship in Melanesia, Bartlett initially wanted to become an anthropologist rather than a psychologist but Rivers persuaded Bartlett that a training in psychophysical methods was the best foundation. Due to many factors, such as his need to head up Cambridge psychology during World War I, Bartlett never realized his anthropological ambitions, but a deep interest in the social permeated all of his later work. The importance Bartlett placed on the social determination of remembering is apparent in this summary statement:

The data presented in the first part of this book have repeatedly shown that both the manner and the matter of recall are often predominantly determined by social influences. (Remembering, p. 244)

Bartlett's desire to integrate cognition and culture is another key reason for the continuing interest in his work.

Bartlett Redux

While interest in Bartlett’s memory work has been ongoing since its first publication, there are peaks where the focus, reexamination and reproduction are more intensive. The first wave of increased interest during the 1940s and 50s largely focused upon replication of the empirical work, and the consequent extension and modification of Bartlett's theories. In retrospect, the publication of Gauld & Stephenson’s 1967 paper documenting a failure to replicate Bartlett's serial reproduction results, closed the era of empirical reworkings of Bartlett’s memory experiments. A number of contemporary memory researchers refer to this paper as proof that Bartlett’s findings from the method of repeated reproduction have not been replicated (Schacter, 1996; Roediger, 1996). Yet, in the same year, Neisser reignited interest in Bartlett’s theoretical framework in his groundbreaking Cognitive Psychology, a book that was instrumental in defining and establishing the field it named. Early in that text Neisser states:

The present approach is more closely related to that of Bartlett (1932, 1958) than to any other contemporary psychologist, while its roots are at least as old as the "act psychology" of the nineteenth century. The central assertion is that seeing, hearing, and remembering are all acts of construction (Neisser, 1967, p. 10).

This second wave initiated by Neisser is characterized by the cognitive emphasis of the memory field at that time, and the work was pursued within the context of the field’s intense focus upon information processing models and the consequent representation of semantic memory as computer-like tree structures. During this period Mandler & Johnson (1977) provided a structural analysis of the story grammar of the War of the Ghosts, and gave a more specific and restrictive meaning to Bartlett’s concept of schemata. Another factor that contributed to the heightened attention to Bartlett’s work in the 70s was Bransford and Franks’ (1971, 1972) clear demonstrations of associative rather than literal recall of sentences.

The third increase in attention to Bartlett’s Remembering is ongoing. Remembering came back into print in 1995 and several papers on Bartlett’s memory work have been published in the last ten years. At least three distinct strands in contemporary psychological inquiry contribute to this third wave of interest: the Vygotskian emphasis on cultural and social aspects of cognition spearheaded by members of the Laboratory of Comparative and Human Cognition (Cole, 1996); the neural net models of memory proposed by parallel distributed processing theorists (McClelland, 1995); and, the emphasis on constructive memory in the bitterly waged recovered/false memory wars (Conway, 1997; Schacter, 1995).

In each of these waves of Bartlett reproduction the individual and collective special interests of psychologists have directed their renderings of Bartlett’s work. The theoretical orientations of the researchers who have revisited Bartlett’s work run the gamut: psychoanalytic, Gestalt, information processing, discourse analysis, social, cognitive and anti-cognitive, and evolutionary perspectives are all represented in the crowd of post-Bartlettians. In line with Bartlett’s theorizing, psychologists have omitted details and points that they find incomprehensible or inconsequential from their own theoretical perspective; as in the War of the Ghosts story renditions, these omissions result in transformation of the narrative, and the gaps must be rationalized. The 'outstanding detail' is determined as much by the interests and attitudes of the experimenters as by the original stimulus, in this case Bartlett’s text. A strong tendency to rationalize, to make sense of the material in line with current interests and emphases, is evident in the work of many Bartlett revisitors. This point was made by one of the early reviewers of Remembering : "Quite in accord with Bartlett’s own thesis, each reader of this book will perceive, recall, and evaluate the contents differently." (Stone, 1934, p. 103). In each of the waves of Bartlett revisiting that I have identified Stone’s thesis is amply confirmed. Each revisiting produces a metonymic rendering of Bartlett’s Remembering formed through an interaction of the individual researcher’s interests and predelictions with the current intellectual climate and preoccupations.

First Wave

In a series of theoretical review papers Oldfield & Zangwill (1942, 1943a, 1943b, 1943c), both former Ph.D. students of Bartletts, subjected the schema concept to intense scrutiny. Northway (1936, 1940a, 1940b) extended Bartlett’s experimental work to a different population in her work with elementary school children, and thoroughly examined the concept of ‘schema’. Allport & Postman (1947) applied Bartlett's methods to the topic of rumor and introduced the Gestalt inspired terminology of 'sharpening' and 'leveling' to the constructive process. Gomulicki (1956), under the direction of Zangwill, studied immediate reproduction of short prose passages and emphasized the abstractive processes that accounted for the selectivity of omissions. Kay (1955) examined the longievity of subject’s versions of two prose passages under conditions of repeated exposure to the original passage. Paul (1959) examined serial reproductions of Bartlett’s version of The War of the Ghosts , plus a more coherent version of the same story. Johnson (1962) replicated Bartlett’s method of repeated reproduction with The War of the Ghosts, and analyzed his results in terms of the Gestalt categories of ‘sharpening’ and ‘normalization’. Gauld & Stephenson (1967) repeated Bartlett’s serial reproduction experiment with more stringent subject instructions, and found fewer errors under these conditions.

Affective Determination: Paul (1959)

A clear example of a metonymic rendering of Bartlett’s theory is found in one of the earliest replications of Bartlett’s serial reproduction experiments. Irving Paul, a psychoanalytically trained psychologist, published his reworking of Bartlett’s experiments in a psychoanalytically oriented journal (Paul, 1959). Paul emphasizes and elaborates upon Bartlett’s focus on the affective realm:

One of Bartlett’s major theses was that cognitive functioning cannot be understood unless it is studied in the light of the subject’s interests, attitudes, affects and goals. (p.3)

When Paul describes the concept of schema he highlights the affective component,

A schema is an observation, simplification, and articulation of experience; part and parcel of its formation and operation are the affective aspects of the experience.

Paul finds that Bartlett’s emphasis on individual subjects’ memories and the particulars of each subject’s background and approach to the task resonates with his clinical experience and training. In keeping with his psychoanalytic background, Paul extends the analysis initiated by Bartlett to produce a typology of rememberers; he divides people into importers and skeletonizers and discusses this categorization in terms of cognitive style.

This focus on individual motivation is, of course, present in Bartlett’s work. In his summary at the end of the chapter on Repeated Reproduction five of the fourteen points ( 7,8,9,11 and 13) address the subjects’ attitudes, interests and affects. He states:

[The rationalizing process] tends to possess characteristics peculiar to the work of the individual who effects it and due directly to his particular temperament and character. (Bartlett, 1932, p. 94)

It is the case, however, that in Paul’s account the emphasis on affective determination and personality characteristics leads to a neglect of other key points in Bartlett’s work, yielding a different theory of memory. The selection and elaboration of the affective and personality characteristics and consequent downplaying of cognitive schematization and social processes in Paul’s work is indicative of the interest driven ‘effort after meaning’ described by Bartlett. This process can be viewed as distorting, but Bartlett’s point is that it is natural and necessary; every reproduction of an original must be filtered through the author’s own perspective, motivations, and current goals.

Paul provided an extensive replication of Bartlett’s serial reproduction experiments. He produced a rationalized version of Bartlett’s rendition of The War of the Ghosts, and found that more coherent versions of the story were more readily and accurately remembered. By providing a story more familiar in context to his subjects, an office tale for secretaries, Paul was able to demonstrate that memory for detail was, as Bartlett stated, determined by preexisting knowledge. Although Paul’s work was a clear replication and extension of Bartlett’s findings, others have reported difficulty reproducing Bartlett’s results.

Failure to Replicate?

[I]n some ways a measure of Bartlett’s stature is that nobody seriously questions the factual results of his experiments. His observations can be repeated, and have been very widely. (Broadbent, 1970, p. 3)

It is curious that Bartlett’s (1932) landmark experiments have never been replicated; see Gauld and Stephenson (1967). (Schacter, 1996, p.320n5)

Clearly, there is some dissent on this issue. As one of the teachers of experimental psychology who routinely employs Bartlett’s War of the Ghosts

Repeated Reproduction experiment as a class assignment, I was puzzled when I first read Schacter’s statement that Bartlett’s results have not been replicated. In reading further I found that many other modern sources concur with Schacter (Alba & Hasher, 1983; Kintsch, 1995; Roediger, Wheeler & Rajaram, 1993; Roediger, 1996). Tracking down the provenance of this idea proved to be an illuminating study in the transmission of the gist of experimental results that again bears out Bartlett’s characterization of the reconstructive transformations that occur in the trafficking of knowledge.

Gauld and Stephenson is the source cited by many as an instance of the failure to replicate, but their experiments differed from Bartlett’s in important ways. Their paper is usually referred to as though Gauld and Stephenson employed the repeated reproduction method For example, Roediger, Wheeler and Rajaram state

Interestingly, we can cite only one attempted replication of Bartlett's (1932) pioneering research --- one published by Gauld and Stephenson (1967) and discussed below --- that tried to confirm his claims using the repeated-reproduction technique. (Roediger, Wheeler & Rajaram, 1993, p. 101)

Yet, Gauld and Stephenson only employed the serial reproduction version of Bartlett’s experiment, with radical alterations to the subject instructions. Gauld and Stephenson’s major focus was the veridicality of memory so they entreated some of their subjects at length to produce the most accurate and literal reproductions possible, using statements such as 'I want you to look on this not as a test of memory, but as an exercise in being as scrupulous and honest as you can in deciding what is and isn’t in the original story.' (Gauld & Stephenson, 1967, p. 41).

In calculating errors Gauld and Stephenson did not include many of the changes discussed by Bartlett: most importantly, omissions were not penalized, synonyms were accepted, general words were accepted in place of particular ones, time order errors were ignored, and place name errors were not counted. In Gauld and Stephenson’s words, "We were concerned with falsehood, and the ‘error rate’ was designed to indicate the proportion of false or untrue statements in the reproductions as a whole." (p. 42). With the more stringent set of instructions subjects did produce a significantly lower error rate (2.78) than when they were simply asked to be as accurate as possible (4.68).

When each subject’s reproduction was complete the experimenter read it back to him or her and asked if there were any phrases he or she was doubtful about. When these doubted phrases were excluded from the analysis the error rate dropped by 42%. Unfortunately, given the form of the data presentation, it is impossible to tell how many of the doubted phrases were in fact correct, but it is fair to say that subjects could not reliably pick out all of their own errors.

Gauld and Stephenson clearly have quite a different conception of the nature of memory than Bartlett. Bartlett emphasizes that remembering is ‘hardly ever really exact’ and that ‘it is not at all important that it should be so’, whereas Gauld and Stephenson focus completely on accuracy. They cast the issue in moral terms:

If a subject does not have a good memory, is not under moral pressure, and is not conscientious, the urge to tell a story, and to be particular rather than general will have its way, and he will probably distort the original passage markedly. (Gauld & Stephenson, 1967, p. 48)

Gauld and Stephenson depict Bartlett as a sloppy experimenter who is leading his unscrupulous subjects with their poor memories astray by not exerting enough moral pressure on them to be honest and accurate, thus they "guess" and "romance". As Carruthers details in her study of memory in medieval culture, morality and memory have long been linked:

Ancient and medieval people reserved their awe for memory. Their greatest geniuses they describe as people of superior memories, they boast unashamedly of their prowess in that faculty, and they regard it as a mark of superior moral character as well as intellect. (Carruthers, 1990, p. 1)

Interestingly, Bartlett's view of memory may have been much closer to medieval conceptions than Gauld and Stephenson's, with their overarching concern for veridicality. In a later book on meditation, rhetoric and imagery Carruthers draws the reader's attention to the differences between modern conceptions of memory and the monastic art of mneme:

So I must ask of my readers a considerable effort of imagination throughout this study, to conceive of memory not only as "rote", the ability to reproduce something (whether a text, a formula, a list of items, an incident) but as the matrix of a reminiscing cogitation, shuffling and collating "things" stored in a random-access memory scheme, or set of schemes -- a memory architecture and a library built up during one's lifetime with the express intention that it be used inventively. (Carruthers, 1998, p. 4)

This emphasis on schemes that organize past experience for the purpose of future uses of the remembered material is consistent with Bartlett's view of memory and will be discussed at length below.

In many ways this controversy typifies the ongoing laboratory/everyday memory debate that provides Bartlett’s point of departure (chapter 1 of Remembering) and has raged ever since. Gauld and Stephenson produce a situation that is quite remote from the normal constraints on memory. In everyday exchanges of narratives or information, or in autobiographical remembering, there is no-one entreating the rememberer to be scrupulous and ultra-conservative. Further, Gauld and Stephenson’s claim that subjects are able to detect their own inaccuracies contravenes the whole eyewitness memory body of work (e.g., Loftus & Loftus, 1980) and current views of false memory (e.g., Conway, 1997).

The discrepancy between Gauld and Stephenson and Bartlett highlights that the everyday/laboratory memory debate is not so much about location, or even stimulus materials, as it is about the researchers’ conception of what memory is for (Baddeley, 1988). Bartlett’s and Gauld & Stephenson’s different conceptions of the purposes of remembering lead to differences in the criteria they set for ‘good memory’; for Gauld & Stephenson only accurate (though not verbatim) reproduction will do, whereas in Bartlett’s view the less literally accurate, and the more transformed for the rememberer’s own purposes, the better for his study of reconstructive processes. Bartlett’s lasting contribution to the everyday/laboratory debate is not that he used more realistic materials. Bartlett knew that a written version of an oral folk tale was a poor representation of the original story (Bartlett, 1923, p. 62), and that the laboratory situation he produced was clearly artificial. What is more radical is his loosening of the literal reproduction criterion of good memory that dominated experimental psychology at the time, and his consequent focus on individual interests driving the development of memory expertise.

Second Wave

Schema

In the cognitive era that superseded behaviorism in American psychology during the 1960s and 70s, Bartlett’s concept of the schema found a new audience. Bartlett introduced his notion of memory schemata in contrast to the prevalent trace theory of memory adhered to by Ebbinghaus and the behaviorist psychologists so dominant at the time Bartlett was writing Remembering. Bartlett and Ebbinghaus differed fundamentally in the metaphor of memory each favored. Ebbinghaus makes his underlying metaphor clear in the following passage from Memory (1885/1964):

These relations can be described figuratively by speaking of the series as being more or less deeply engraved in some mental substratum. To carry out this figure: as the number of repetitions increases, the series are engraved more and more deeply and indelibly; if the number of repetitions is small, the inscription is but surface deep and only fleeting glimpses of the tracery can be caught; with a somewhat greater number the inscription can, for a time at least, be read at will; as the number of repetitions is still further increased, the deeply cut picture of the series fades out only after ever longer intervals. (Ebbinghaus, 1885/1964, pp. 52-3)

This engraving metaphor version of a trace theory was consistent with the stimulus-response theories of behaviorist psychologists. Bartlett called this trace theory "lifeless", and instead emphasized the active role of a person's attitudes, interests, appetites, feelings and values in organizing the material to be remembered. This emphasis on the individual's activity demonstrates James Ward's profound influence on Bartlett's thinking. In his obituary of Ward Bartlett (1925) drew attention to the effect on Ward of Lötze's 'constant reiteration of the fundamental importance of life and activity' (Bartlett, 1925, p. 450). Bartlett conceived of a person's past experiences acting as an organized mass, rather than as individual records of events that differed only in strength.

…in all relatively simple cases of determination by past experiences and reactions, the past operates as an organized mass rather than as a group of elements each of which retains its specific character. (Remembering, p. 197)

While the history of Bartlett's emphasis on the active organizational role played by schemata can be traced through Ward to Lötze, his inspiration for employing the term schema was drawn directly from neurologist Henry Head. Head & Holmes (1920) used 'schema' to refer to the unconscious representation of body posture that they describe more fully here:

It would be impossible to discover the position of any part of the body, unless the immediate postural impressions were related to something that had preceded them. A direct perception of posture, analogous to that of roughness, cannot occur; in every case, the new position of the limb is related to some previous posture. We have been able to show that the standard against which a change in posture is estimated is not an image either visual or motor; it lies outside consciousness. Every recognizable change in posture enters consciousness already charged with its relation to something which has gone before, and the final product is directly perceived as a measured postural change. For this combined standard, against which all subsequent changes in posture are estimated, before they enter consciousness, we have proposed the word 'schema'. (Head & Holmes, 1920; cited in Brain, )

Bartlett employed this notion of a combined standard built on the basis of past experience to capture his view that 'the past operates as an organized mass'. Following Head, Bartlett used a motor example to illustrate the concept:

Suppose I am making a stroke in a quick game, such as tennis or cricket. … When I make the stroke I do not, as a matter of fact, produce something absolutely new, and I never merely repeat something old. The stroke is literally manufactured out of the living visual and postural 'schemata' of the moment and their interrelations. I may say, I may think that I reproduce exactly a series of text-book movements, but demonstrably I do not; just as, under some other circumstances, I may say and think that I reproduce exactly some isolated event which I want to remember, and again demonstrably I do not. (Remembering, 1932, p. 201-202.

The schema concept allowed Bartlett to set up an alternative to trace theory and thereby escape the classic view of memory as the re-excitement of individual traces of past experiences. Bartlett was not wholly satisfied with the schema nomenclature and his alternative suggestions for terminology - active developing patterns and organized setting - reveal that he emphasized the plastic nature of schemata in applying Head's concept to memory. Yet, Bartlett did not simply apply Head's concept to a new domain, rather he extended and transformed it in the process.

Two related terms - attitude and interest - are crucial to understanding Bartlett's reworking of the schema concept. Bartlett notes that the first thing to emerge in any act of remembering is often an attitude which is 'largely a matter of feeling or affect' (Remembering, 1932, p. 207). It's important to note that this 'feeling or affect' is a wider concept that emotion, and that it is a significant part of Bartlett's phenomenology of memory. Recall is viewed as a construction made on the basis of this attitude that serves to justify it. Zangwill (1972) traces Bartlett's use of the concept of attitude to Betz's Einstellung, translated as mental posture or set. In his extensive review of emotions and memory psychoanalyst David Rapoport locates Bartlett's use of the concept of attitude in the context of German introspectionist psychology and equates it with 'selective force' (Rapoport, 1941). Larsen & Berntsen (2000) draw attention to the fact that the notion of attitude has led a quiet life in comparison to the substantial influence of the concept of schema. They argue that this relative neglect is attributable to the mentalistic nature of attitude. As Ross (1991) notes in his history of the study of autobiographical memory, when Bartlett was writing Remembering much of the earlier subjective approach lingered but this aspect of Bartlett's thinking about memory has been 'tactfully ignored' (Ross, 1991, p. 21). Discomfort with this subjective aspect of Bartlett's work plays in to Zangwill's (1972) dismissal of the significance of the schema concept. In their thorough series of theoretical reviews of Head, Bartlett and Wolters' use of schemata, Oldfield and Zangwill (1942-3) make it is clear that they wished to discard some of the more mentalistic and subjective aspects of Bartlett's thinking, such as his equation of consciousness with 'turning around on one's own schemata'. In a later retrospective paper Zangwill (1972) is more explicit about his despair about the coherence and fruitfulness of Bartlett's chief theoretical concept.

Northway (1940a) also emphasizes the multiple ways that Bartlett uses the term 'schema'. She attributes the variation and inconsistency in Bartlett's application of the term to his attempt to integrate Ward's psychology with Head's physiology and Rivers' anthropology. Northway wishes instead to redefine schema to bring it in line with the active view of memory:

To Lötze and all idealistic psychologists the memory is a crucial instance; there is no faculty of memory, no cerebral organ, no ‘storehouse of ideas’; only the living continuity of the soul and its power of reproducing its previous activities ... Memory does not keep any picture; what it really retains is a kind of schema, a plan of action. (Brett, 1921, cited in Northway, 1940)

Northway's definition of schema is 'what the subject makes from the given material'. She explores this specific use of the term through experimental work on children's learning and creation of stories and poetry, arguing for a progressive view of education on the basis of a more constructive and creative view of memory. I suggest that due to the general and somewhat vague definition of schema provided by Bartlett all later investigators of the concept must first narrow their application of the term. This is evident in Mandler and Johnson's second wave study of The War of the Ghosts. Mandler & Johnson (1977) reexamined the structure of Bartlett’s most frequently employed folktale in the context of story grammars, and in so doing produced a new metonymic rendering of Remembering, that emphasized structure at the expense of content. Their point of departure in this reanalysis of Bartlett’s work was the somewhat incomprehensible nature of the folktale used by Bartlett.

A Bizarre and Disjointed Tale: Mandler & Johnson (1977)

The War of the Ghosts bears about as much resemblance to normal prose as nonsense syllables do to words. In fact, much of Bartlett’s analysis of recall of this passage is about the mental gymnastics his English college students had to use in comprehending and remembering such a bizarre and disjointed tale. (Roediger, 1997, p. 492)

Much of the subsequent criticism of Bartlett’s work focuses around this point: that the War of the Ghosts story is exceptional. Mandler and Johnson (1977) provide a particularly clear example of the cognitivist transformation of Bartlett’s work. They perform a story grammar analysis of Bartlett’s version of The War of the Ghosts. As Mandler and Johnson themselves acknowledge, their application of the schema concept in this context is much more specific and narrow than Bartlett’s. They redefine Bartlett’s schema as a ‘story schema’, by which they mean "an idealized internal representation of the parts of a typical story and the relationship among these parts" (p. 111), and "a set of expectations about the internal structure of stories which serves to facilitate both encoding and retrieval." (p.112). Thus, their focus is exclusively upon structural characteristics of the story rather than specific content. Mandler and Johnson identified several points where The War of the Ghosts violated the rules of the grammar they developed for well formed folktales. They found that "the predominant characteristic of Bartlett’s version of the story is the presence of temporal relations where causal ones are expected" (p. 138). By reanalyzing the reproductions provided by Bartlett they demonstrated that their story grammar predicted the points at which subjects would have most trouble: 5 of the 6 propositions that almost no-one recalled were next to a violation in the story grammar and distortions in content also occurred most frequently at these points.

This exclusive focus upon the structural considerations leads to a neglect of many of the factors Bartlett deemed most important: individual special interests, affect and temperament, and social determination of recall. Mandler and Johnson’s concept of a story schema could be coded and form the basis of a ‘story reproduction’ program. Mandler and Johnson argue that Bartlett’s subjects were forced to forget or invent at predetermined points in the story because of its faulty structure. This analysis fits nicely with the information processing paradigm dominant in cognitive psychology at the time of Mandler and Johnson’s publication, but it leaves by the wayside the enormous individual variation in story reproductions that so impressed Bartlett. Mandler and Johnson’s rendition of Bartlett is metonymic; by substituting part of Bartlett’s meaning of schema for the whole they transformed the concept to bring it in line with their theoretical perspective.

The issue of programming representations of knowledge, such as schemata, was actively pursued in the cognitive science and Artificial Intelligence communities in the 1970s (Bransford, 1979). Not all researchers referred to them as schemata, the terms ‘frames’ and ‘scripts’ were used for the same purposes: to precisely specify knowledge of the world that could be used to comprehend new information. A compelling point that these studies quickly unearthed was the extreme difficulty of capturing the extent of human knowledge, even when quite simple concepts or restricted situations are employed. Bransford gives the example of the synonyms ‘catch’ and apprehend’: we can say "The policeman apprehended the thief", but not "The policeman apprehended the ball". In the case of a computer program everything has to be specified, and that turns out to be a Sisyphean task. The strategy adopted during this period of interest in Bartlett's concept of schema was to tightly specify the domain of application, working with schemata or scripts for commonly occurring events, such as the restaurant script. These scripts were then used to comprehend inputs in the form of stories (Schank, 1977). The striking thing about the computer rewrites of simple restaurant tales is their banality, they contain none of the imaginative rationalizing power that Bartlett noted in his subjects’ reproductions. Bartlett’s subjects alteredThe War of the Ghosts by adding statements about the characters’ thoughts, motivations, intentions and feelings, whereas the computer rewrites added many more descriptive behavioral steps without enriching the ‘theory of mind’ content of the stories. Reviewing the AI script work of this 'second wave' period reinforces the point that applying the notion of schemata is a selective process that results in transformation of Bartlett’s concept.

The uses of schema in current cognitive psychology are manifold. Studies of expertise in the Newell and Simon tradition emphasize the development of memory skills involving 'chunking' or the building of schemata (Chase & Simon, 1973; Gobet & Simon, 1998). Research on the development of children's memory has employed a version of the concept of schema as a general knowledge base (Nelson, 1993). The pattern detecting aspect of schema has been expanded in parallel distributed processing models of memory (McClelland, 1995). The flexibility of the term schema has led to redefinitions of Bartlett’s concept that depart quite radically from his original intent. As Bartlett feared, the term is ‘at once too definite and too sketchy’ (Remembering, p. 201). The looseness of Bartlett’s concept allowed for a stricter and narrower information processing interpretation in subsequent cognitive psychology and artificial intelligence circles.

Third Wave

The third marked wave of interest in Bartlett’s work is motivated by the wide-ranging and unified view of mental life afforded by his theories; few influential experimental psychologists maintained the serious interest in the social formation of cognition evinced by Bartlett. Bartlett was much concerned with social influences on personal recall. Remembering is subtitled ‘A Study in Experimental and Social Psychology’, and the second half of the book is entitled ‘Social Psychology’. The lack of attention paid to the second section of the book in the first and second waves of Bartlett reproduction is what led Edwards and Middleton to suggest that it often seems as if researchers are referencing two different books, "one on cognitive aspects of individual recall, and another on social factors influencing recall" (Edwards & Middleton, 1987, p. 78). The third wave of Bartlett revival pays much closer attention to his social psychology; here, we see the barometric function of Bartlett’s Remembering in action.

Giving Up the Social Ghost

Ironically, the beginnings of the renewed focus on social aspects of Bartlett's work was sparked by an outright denunciation of the capabilities of psychologists to seriously attend to the social nature of humans. In her biography of Evans-Pritchard, anthropologist Mary Douglas flatly states: "The author of the best book on remembering forgot his own first convictions. He became absorbed into the institutional framework of Cambridge University psychology and restricted by the conditions of the experimental laboratory." (p.19). In her later book How Institutions Think she is even more declarative and emphatically anti-psychological:

According to the thought style of his day, it was improbable that institutional constraints would have much influence on moderns, so it would not be worth searching for them. … His career is a self-referencing instance of the claim that psychologists are institutionally incapable of remembering that humans are social beings. As soon as they know it, they forget it. (p. 81)

Douglas’s rendition of Bartlett’s view is a classic instance of omission and condensation transforming the meaning of the whole that typifies the treatment of Bartlett’s work. She states, "Bartlett thought that a strong social support to memory was inclined to produce a mechanical style of recall." While Bartlett does, indeed, contrast the automatic character of a Swazi herdsman’s recall of cattle bought a year earlier with the effortful, constructive recall of a map by a geologist and mining expert a year after prospecting a site, he is not slavishly equating social support with mechanical recall. He goes on to state,

Now of course it would be absurd to maintain that the socially determined recall is always predominantly of the first, recitative type; while remembering which is directed by individual interests is predominantly of the second, the constructive type. Indeed I shall later urge that social determination of recall often affords just the basis for that constructiveness which has been found already to characterize many instances of recall. (pp. 251-2)

These are not the words of a psychologist who has forgotten that humans are social beings. Further, in the results of his experimental work, which Douglas deemed incapable of capturing institutional effects, Bartlett discerns clear social influences on remembering:

A glance back over the chapter on Repeated Reproduction will give us numerous cases. The ‘old mother at home’ and the ‘filial piety’ were both direct expressions of family group influences. The occasion and directness of rationalization were alike found to be largely given by social conditions. Much outstanding detail owed its prominence to social influences. (pp.252-253)

When you further consider that one of Bartlett’s next books was on Political Propaganda, a clear instance of institutional constraints upon moderns, it becomes clear that Douglas’s anthropological focus has led her to underestimate the strong social thread in Bartlett’s work.

Costall, a psychologist noted for his anti-cognitivist stance, produced two scholarly articles on the development of Bartlett's thought (Costall, 1991, 1992). He concurs with Douglas that Bartlett abandoned his early radically social conception of cognition for an individualistic and mechanical view inspired by the cybernetic approach and his WWII driven work on human-machine interfaces. Although Bartlett’s extensive work in applied psychology was concerned with the impact of human perception and cognition on the use of machines he did not view humans as machine like. In his obituary of his former student, Kenneth Craik, Bartlett makes this clear:

He seemed to be trying to see them as evidence that in so far as they are successful, they show how the mind works, not in inventing the machines and using them, but in actually solving the problems. … Both of these inferences are dubious. (Bartlett, 1946, p. 114)

Other modern reexaminers of Bartlett's work have taken the opposite tack of forefronting and stressing the importance of Bartlett's characterization of memory as socially and culturally influenced. For example, Edwards and Middelton (1987) explicitly recast Bartlett's work to illustrate its compatibility with their focus on discourse analysis.

Talking with the Subject: Edwards & Middleton (1987)

Edwards and Middleton bemoan the one-sided representation of Bartlett as a major forerunner of the information processing approach to memory and develop the argument that a previously hidden thread in Bartlett’s work is the key to restoring it to its proper place as the foundation of a ‘broader, culturally contextualized and functional approach to the study of everyday remembering’ (Edwards & Middleton, 1987, p. 77). Given Edwards and Middleton’s commitment to the methodology of discourse analysis, it is natural that they perceive the previously ignored Bartlettian emphasis on conversational discourse. While Edwards and Middleton do emphasize an aspect of Bartlett's writings that accords best with their own theoretical standpoint, their rendering of Bartlett is not as restrictive as some previous accounts because they point to other neglected aspects of Bartlett's work: the role of 'feeling' and 'attitude' in remembering, and the cross-modal nature of 'symbolic remembering'.

As Edwards and Middleton illustrate (pp. 85-87), Bartlett’s text is replete with accounts of conversations with his subjects. From reading Bartlett’s texts, including his many accounts of the development of the Cambridge school of psychology, and his students’ published appreciations, I get the impression that Bartlett was a man who loved to talk about psychology with his students, his friends, and his subjects (I suspect these were widely overlapping categories). His genuine interest in what his subjects have to say about their approach to his tasks is communicated throughout his discussions of experimental results. In his autobiography, which is really a personal synopsis of his conception of psychology, Bartlett makes his position on first person accounts clear: "Very often the most valuable information can be given in terms possible only to the person himself who responds." (Bartlett, 1936, p.42). This position was very much against the grain of the then current behavioristic psychology and the later information processing psychology, both of which focus on the lawfulness and generalities of subjects’ responses, rather than their idiosyncrasies and differences.

Although the subjective reports of subjects’ are taken seriously, and plenty of discourse is reported, Bartlett does not analyze the pragmatics of the experimental situation in the manner characteristic of discursive psychologists. For example, in their reanalysis of Neisser’s John Dean case study, Edwards and Potter focus on the pragmatic function of the statements Dean makes about his own memory (Edwards & Potter, 1992). In contrast, Bartlett takes his subjects’ statements about their memory at face value and never probes their communicative intent. In fact, his seeming unawareness of the pragmatics of the situation when he tests the only non-British subject, an Indian he describes as "impressionable, imaginative, and, using the word in its ordinary conventional sense, nervous to a high degree" (Remembering, p.75-76), is jarring from a contemporary viewpoint. Bartlett seems unaware of the role that his own presence and stature might have in increasing the nervousness of the colonized subject.

Edwards and Middleton document the prevalence and importance of conversational exchange in Bartlett's theories of remembering, but the theoretical significance they attach to this finding diverges radically from Bartlett's viewpoint. In their self aware reinvention of Bartlett Edwards and Middleton develop a theme that becomes familiar when examining recent writings about Bartlett's Remembering: the need to augment Bartlett's theorizing with other conceptions of the social nature of recall. Many recent papers suggest the fusion of Bartlett’s views with those of other theorists such as Bahktin (Beals, 1998), Wittgenstein (Shotter, 1990), Moscovici (Saito, 1996), and Edelman (Saito, 1996).

Augmenting Bartlett's Conception of the Social

Beals (1998) argues that Bartlett’s conception of the social is inadequate and a marriage with the Russian literary theorist Bakhtin’s perspective would remedy this deficiency. Bakhtin shared Vygotsky’s emphasis on the social formation of mind through language (Vygotsky, 1962). Nowhere in Bartlett's writing is there any emphasis on the learning of language as the means whereby the social constitutes individual cognition. The concepts of internalization and inner speech, so crucial to Vygotsky and Bakhtin’s theorizing, are absent from Bartlett’s writings. Perhaps this lack of emphasis on the social nature of language derives from the fact that Bartlett’s chief analogy for schemata was bodily rather than verbal. In light of the wealth of recent work on the development of 'socially accessible memory' (Pillemer & White, 1989; Pillemer, 1998) the absence of this theme from Bartlett's theorizing is significant. Attention to the work of Vygotsky more naturally leads to the current emphasis on the importance of the extent and form of memory talk in the construction of personal event memory.

Wittgenstein was another contemporary of Bartlett’s who paid serious attention to the interrelation of the mental, linguistic, and social (Wittgenstein, 1953). Although Bartlett and Wittgenstein were both luminaries in Cambridge, England between the wars, I can find no reference to any mutual awareness. Consequently, Wittgenstein’s penetrating account of the social nature of language and its role in the formation of our mentality, does not inform Bartlett’s work. Yet, there are some striking congruencies in their approaches. A Wittgensteinian analysis of remembering has the same point of departure as Bartlett took in Remembering: a dismantling of the notion that memories consist of faint traces or images of isolated past events (Malcolm, 1977; Shotter 1987, 1990). Bartlett’s focus on the everyday uses of memory and the way that enduring individual interests generate memory expertise fits well with Wittgenstein’s central notion of forms of life. While Bartlett’s conception of the social nature of remembering was not fully developed along the lines suggested by Wittgenstein, Vygotsky, and Bakhtin, it was a strong and integral aspect of his approach.

False Memories of Bartlett

At least some of the renewed interest in Bartlett’s work can be attributed to the current furor over false memories. While Bartlett’s text is not referenced in any of the trade publications directly addressing questions of false memories and sexual abuse, written mainly by clinicians and journalists, the publications stemming from the experimental psychology of memory field are full of references to Bartlett as the originator of the view of memory as constructive and imaginative (Conway, 1997; Schacter, 1995). Bartlett’s characterization of memory does provide a natural and compelling explanation of memory illusions, something I can illustrate through a personal example.

I developed a flagrantly false memory in connection with my studies of Bartlett’s work. I was searching for an encyclopedia entry Bartlett wrote on the topic of 'Memory' which Costall (1990) cites as evidence of Bartlett's later turn towards an information processing view of memory. It was not available in either of the editions of Encyclopaedia Britannica owned by my college library, so I asked my parents to send me a copy of the article that I remembered reading in my family's copy of EB. I have a clear memory of finding the article and being delighted that it was written by Bartlett; I remember exactly where I was sitting (I can imagine the squishy feel of the foam cushion) and I have a clear image of my surroundings. After causing much familial confusion, I realized that I had fabricated this pleasing memory. My family owns the 1976 edition, by which time Zangwill’s article, which contains no mention of Remembering, had replaced Bartlett’s.

My image of finding and reading Bartlett’s article at home is still clear and specific, but it could not have occurred. In Bartlett’s words, "the appearance of a visual image is followed by an increase in confidence entirely out of proportion to any objective accuracy that it thereby secured." The image is not based on nothing: I sat in that living room, in that very chair, reading the encyclopedia many times, I had looked at the entry under Memory in the past, but my current fascination with Bartlett led me to unwittingly transfer my independent knowledge that Bartlett wrote an EB article on Memory to construct a specific memory of an event that never occurred, but was plausible given my various pieces of relevant knowledge at that time. In constructing this memory I wove together pieces from various ‘schemata’ to make a coherent whole that formed a convincing autobiographical memory.

In order to construct this memory I had to omit several other pieces of knowledge: that Zangwill had written the article, that there was nothing about Remembering in the article, that the timing was not correct, and I even had to override an uneasiness about the slight familiarity of Zangwill’s article when I looked at it in the Sarah Lawrence library, though that probably had more to do with the layout and feel of the pages than the content.

The argument of this paper is that the various constructions of Bartlett’s views on memory share this potentially distorting quality: the selection of outstanding detail in line with the scholars’ interests leads inevitably to the omission of other aspects of Bartlett’s eclectic work with its wide theoretical net. Due to the interconnected nature of the many threads of Bartlett’s position, these omissions will lead to transformations in the view of remembering expounded. Bartlett recognized that omissions often form the basis for transformations:

In a story series of this type, any omission from an individual version is liable to become significant and to account for a succession of connected changes in subsequent versions. (Remembering, p.125)

One clear omission in light of the recruitment of Bartlett’s views within the false memory literature is his concomitant emphasis on stereotypy. Although transformations and importations occur there is also a strong tendency for memories to become quickly fixed, especially with frequent reproductions. Kay (1955) demonstrated this most compellingly in his reworking of Bartlett’s repeated reproduction experiments. He modified Bartlett’s technique by ending each session with a re-reading of the original passage. Remarkably he found that successive versions bore a much greater resemblance to their immediate predecessors than to the original passage. Despite their reexposure to the original passage, his subjects were "repeating the same correct items, making the same errors, and omitting the same items one week after another" (Kay, 1955, p. 89). As Kay succinctly expresses his major finding, "They could be consistent with themselves, but not with the original." (Kay. 1955, p. 92).

One of the folk psychological beliefs about memory, that is instituted in court cross-examination procedures, is that if someone is giving a true account then they will give consistent versions of their story on different occasions, and that inconsistencies unmask lying and falsehood. In fact, Bartlett’s work undermines this view of memory. The concept of ‘stereotypy’ means that once a detail has entered someone’s narrative, it is very likely to remain, especially if the story is told frequently. Further, Bartlett suggests that it is the imported or false portions of the story that the teller may be most confident about, particularly if they play a significant role in making the narrative coherent or generate a compelling mental image.

Reproductions Yet to Come

Although recent revisitors of Bartlett’s work have been more explicit about their reconstructive purpose than the many earlier reinventors of the Bartlettian tradition, the tendency to omit, and thereby alter the shape of Bartlett’s legacy, continues. If Bartlett is right about the nature of human cognition, it could not be otherwise. One important thread that has not been stressed in recent employment of Bartlett’s work is his intense focus on the role of individual interests in motivating remembering. Another radical aspect of Bartlett’s work is his conception of the role of mental imagery. Bartlett successfully dismantled the idea that memories were like faint traces or images: literal, if faded, reproductions of single past events or circumstances. Instead he considered images as the means of selecting from the mass of past experiences:

In general, images are a device for picking bits out of schemes, for increasing the chance of variability in the reconstruction of past stimuli and situations, for surmounting the chronology of presentations. By the aid of the image, and particularly of the visual image ... a man can take out of its setting something that happened a year ago, reinstate it with much if not all of its individuality unimpaired, combine it with something that happened yesterday, and use them both to help him to solve a problem with which he is confronted to-day. (Remembering, p. 219)

This view of memory as the imaginative combination of many past experiences brought to consciousness by the present needs and purposes of the individual is compatible with modern conceptions of memory processes (e.g., Schacter, 1996), and draws our attention back to the prominent role of imagery in the experience of remembering.

In closing his affectionate biographical memoir of Bartlett, Donald Broadbent states, "I myself feel that some of Bartlett’s insights have still not had their full impact, and that they will come into their own in the next generation" (Broadbent, 1970, p. 8). Broadbent’s statement was prophetic, interest in hitherto neglected aspects of Bartlett’s work has been lively in recent years. I am glad that one of the early reviewers of Remembering has been proved wrong by time:

The book will find a place upon the shelves of those who study remembering, but it will not be in the special section reserved for those investigators whose writings have become landmarks in the advance towards the comprehension of this important problem. (Jenkins, 1933, p. 715)

Bartlett’s engaging, lively, and humane text has a lasting place on the shelves of memory researchers and many insights yet to yield.

Bibliography

Alba, J.W. & Hasher, L. (1983) Is Memory Schematic? Psychological Bulletin, 93, 203-231.

Allport, G.W. & Postman, L. (1947) The Psychology of Rumor, New York: Henry Holt.

Baddeley, A. (1988) But what the hell is it for? In Gruneberg, M.M. & Sykes, R.N. (Eds.) Practical Aspects of Memory, Vol.1, Wiley: New York..

Banaji, M.R. & Crowder, R.C. (1989) The bankruptcy of everyday memory, American Psychologist, 44, 1185-1193.

Bartlett, F.C. (1923) Psychology and Primitive Culture, Cambridge University Press: New York.

Bartlett, F.C. (1932) Remembering: A Study in Experimental and Social Psychology, Cambridge University Press: New York.

Bartlett, F.C. (1936) Autobiography, in C. Murchison (ed.), A History of Psychology in Autobiography, vol. 3, Worcester, MA: Clark University Press, pp.39-52.

Bartlett, F.C. (1946) Obituary Notice, Kenneth J.W. Craik, 1914-1945, British Journal of Psychology, 36, 109-116.

Bartlett, F.C. (1955) Fifty years of psychology, Occupational Psychology, 29, 203-216.

Bartlett, F.C. (1958) Thinking: A Study in Experimental and Social Psychology, Basic Books: New York.

Beals, D.E. (1998) Reappropriating Schema: Conceptions of Development From Bartlett and Bakhtin, Mind, Culture, and Activity, 5, 3-24.

Boas, F. (1901) Kathlamet Texts, Bureau of American Ethnology, Bulletin 26, 5-260.

Bransford, J. D. (1979) Human Cognition: Learning, Understanding and Remembering, Wadsworth: Belmont, CA.

Bransford, J.D. & Franks, J. (1971) The abstraction of linguistic ideas, Cognitive Psychology, 2, 331-350.

Bransford, J.D. & Franks, J. (1972) The abstraction of linguistic ideas: A review, Cognition, 1, 211-250.

Brewer, W.F. (2000) Bartlett's Concept of the Schema and its Impact on Theories of Knowledge Representation in Contemporary Cognitive Psychology. In Saito, A. (Ed.) Bartlett, Culture & Cognition, Guildford, UK: Psychology Press.

Broadbent, D.E. (1970) Frederic Charles Bartlett 1886-1969, Biographical Memoirs of Fellows of the Royal Society, 16, 1-13.

Carruthers, M. (1990) The Book of Memory, New York: Cambridge University Press.

Carruthers, M. (1998) The Craft of Thought. New York: Cambridge University Press.

Chase, W.G. & Simon, H.A. (1973) Perception in chess. Cognitive Psychology, 4, 55-81.

Cole, M. (1996) Cultural Psychology: A Once & Future Discipline, Cambridge, MA: Harvard University Press.

Conway, M.A. (1997) Recovered Memories and False Memories, Oxford University Press: New York.

Costall, A. (1992) Why British Psychology is Not Social: Frederic Bartlett’s Promotion of the New Academic Discipline, Canadian Psychology, 33, 633-639.

Costall, A. (1991) Frederic Bartlett and the rise of prehistoric psychology, In Against Cognitivism, Still, A. & Costall, A. (Eds.), Harvester Wheatsheaf: New York.

Clark, A. (1997) Being There: Putting Brain, Body, and World Together Again. Cambridge, MA: MIT Press.

Crampton, C. (1978) The Cambridge School: The life, work and influence of James Ward, W.H.R. Rivers, C.S. Myers and Sir Frederic Bartlett, Ph.D. Thesis, University of Edinburgh.

Douglas, M. (1980) Evans-Pritchard, New York: Fontana.

Douglas, M. (1987) How Institutions Think, New York: Routledge.

Ebbinghaus, H. (1885/1964) Memory: A Contribution to Experimental Psychology, (H.A. Ruger & C.E. Bussenius, Trans.) New York: Dover. (original work published 1885).

Edwards, D. & Middleton, D. (1987) Conversation and Remembering: Bartlett Revisited, Applied Cognitive Psychology, 1, 77-92.

Edwards, D. & Middleton, D. (1990) Collective Memory, Sage: New York.

Edwards, D. & Potter, J. (1992) Discursive Psychology, Sage:New York.

Ericsson, K.A. & Kintsch, W. (1995) Long-term working memory, Psychological Review, 102, 211-245.

Gauld, A. & Stephenson, G.M. (1967) Some experiments related to Bartlett's theory of remembering, British Journal of Psychology, 58, 39-49.

Gobet, F. & Simon, H.A. (1998) Expert Chess Memory: Revisiting the Chunking Hypothesis, Memory, 6, 225-255.

Gomulicki, B.R. (1953) The Development and Present Status of the Trace Theory of Memory, Cambridge University Press: New York.

Gomulicki, B.R. (1956) Recall as an Abstractive Process, Acta Psychologia, 12, 77-94.

Hunter, I.M.L. (1957) Memory, Penguin Books: Baltimore, Maryland.

Jenkins, J.G. (1933) Review of Remembering by F.C. Bartlett, Journal of Psychology, 712-715.

Johnson, R.E. (1962) The Retention of Qualitative Changes in Learning, Journal of Verbal Learning and Verbal Behavior, 1, 218-223.

Kay, H. (1955) Learning and Retaining Verbal Material, General Psychology, 46, 81-100.

Kintsch, W. (1995) Introduction to the Reissue of Remembering, In Bartlett, F.C., Remembering, xi-xv.

Loftus, E.F. & Loftus, G.R. (1980) On the permanence of stored information in the human brain, American Psychologist, 35, 409-420.

McGeoch, J.A. (1933) Review of Remembering, Psychological Bulletin, 30, 774-776.

Malcolm, N. (1977) Memory and Mind, Cornell University Press: Ithaca, NY.

Mandler, J.M. & Johnson, N.S. (1977) Remembrance of Things Parsed: Story Structure and Recall, Cognitive Psychology, 9, 111-151.

Murray, D. (1976) Research on Human Memory in the Nineteenth Century, Canadian Journal of Psychology, 30, 201-220.

Neisser, U. (1967) Cognitive Psychology, Appleton Century Crofts: New York.

Neisser, U. (1976) Cognition and Reality: Principles and Implications of Cognitive Psychology. New York: Freeman.

Northway, M.L. (1936) The Influence of Age and Social Group on Children’s Remembering, British Journal of Psychology, 27, 11-29

Northway, M.L. (1940) The Concept of the ‘Schema’: Part 1, British Journal of Psychology, 30, 316-325.

Northway, M.L. (1940) The Concept of the ‘Schema’: Part 2, British Journal of Psychology, 31, 22-36.

Oldfield, R.C. (1954) Memory Mechanisms and the Theory of Schemata, British Journal of Psychology,,14-23.

Oldfield, R.C. & Zangwill, O.L. (1942) Head’s Concept of the Schema and its Application in Contemporary British Psychology, Part I: Head's Concept of the Schema,The British Journal of Psychology, 32, 267-286.

Oldfield, R.C. & Zangwill, O.L. (1943) Head’s Concept of the Schema and its Application in Contemporary British Psychology, Part II: Critical Analysis of Head's Theory,The British Journal of Psychology, 33, 58-64.

Oldfield, R.C. & Zangwill, O.L. (1943) Head’s Concept of the Schema and its Application in Contemporary British Psychology, Part III: Bartlett's Theory of Memory,The British Journal of Psychology, 33, 113-129.

Oldfield, R.C. & Zangwill, O.L. (1943) Head’s Concept of the Schema and its Application in Contemporary British Psychology, Part IV: Wolters' Theory of Thinking,The British Journal of Psychology, 33, 143-149.

Paul, I.H. (1959) Studies in remembering, Psychological Issues, 1, No. 2, 1-152.

Rapaport, D. (1971) Emotions and Memory, New York: International Universities Press.

Roediger III, H.L. (1996) Memory Illusions, Journal of Memory and Language, 35, 76-100.

Roediger III, H. L. (1997) Review of Remembering: A Study in Experimental and Social Psychology, Frederic C. Bartlett, Contemporary Psychology, 42, 488-492.

Roediger III, H.L., Wheeler, M.A. & Rajaram, S. (1993) Remembering, Knowing, and Reconstructing the Past, The Psychology of Learning and Motivation, 30, 97-134.

Ross, B. (1991) Remembering the Personal Past: Descriptions of Autobiographical Memory, New York: Oxford University Press.

Ross, M. (1989) Relation of Implicit Theories to the Construction of Personal Histories, Psychological Review, 96, 341-357.

Saito, A. (1996) Social Origins of Cognition: Bartlett, Evolutionary Perspective and Embodied Mind Approach, Journal for the Theory of Social Behavior,

Saito, A. (2000) Bartlett, Culture & Cognition, Guildford, UK: Psychology Press.

Schacter, D.L. (1995) Memory Distortion: How Minds, Brains, and Societies Reconstruct the Past, Harvard University Press: Cambridge, MA.

Schacter, D. L. (1996) Searching for Memory, New York: Basic Books.

Schank, R. (1977) Scripts, plans, goals, and understanding: An inquiry into human knowledge structures, Lawrence Erlbaum Associates: Hillsdale, NJ.

Shotter, J. (1987) Remembering and Forgetting as Social Institutions, The Quarterly Newsletter of the Laboratory of Comparative Human Cognition, 9, 11-19.

Shotter, J. (1990) The Social Construction of Remembering and Forgetting, In Edwards, D. & Middleton, D. (Eds.) Collective Memory, Sage: New York.

Stone, C.L. (1934) Review of Remembering, Journal of Abnormal and Social Psychology, 29, 103-105.

Thompson, Charles P. (1997) Review of Remembering: A Study in Experimental and Social Psychology, Frederic C. Bartlett, Contemporary Psychology, 42, 492-493.

Vygotsky, L.S. (1978) Mind in Society, Harvard: Cambridge, MA.

Wheeler, M.A. & Roediger, H.L. (1992) Disparate Effects of Repeated Testing: Reconciling Ballard’s (1913) and Bartlett’s (1932) Results, Psychological Science, 3, 240-245.

Wilson, E.A. (1998) Neural Geographies: Feminism and the Microstructure of Cognition, Routledge: New York.

Wittgenstein, L. (1953) Philosophical Investigations, Blackwell: New York.

Wynn, V.E. & Logie, R.H. (1998) The Veracity of Long-Term Memories - Did Bartlett get it Right?, Applied Cognitive Psychology, 12, 1-20.

Zangwill, O. (1972) Remembering Revisited, Quarterly Journal of Experimental Psychology, 24, 123-138.

Appendix

The War of the Ghosts

One night two young men from Egulac went down to the river to hunt seals and while there it became foggy and calm. They heard war cries and they thought, "Maybe this is a war party." They escaped to the shore and hid behind a log. Now canoes came up, and they heard the noise of paddles and saw one canoe coming up to them. There were five men in the canoe, and they said:

"What do you think? We wish to take you along. We are going up the river to make war on the people."

One of the young men said: "I have no arrows."

"Arrows are in the canoe," they said.

"I will not go along. I might be killed. My relatives do not know where I have gone. But you." he said turning to the other, "may go with them."

So one of the young men went, but the other returned home.

And the warriors went on up the river to a town on the other side of Kalama. The people came down to the water, and they began to fight, and many were killed. But presently the young man heard one of the warriors say, "Quick, let us go home; that Indian has been hit."

Now he thought, "Oh, they are ghosts." He did not feel sick,

but they said he had been shot.

So the canoes went back to Egulac, and the young man went ashore to his house and made a fire. And he told everybody and said, "Behold, I accompanied the ghosts, and we went to a fight. Many of our fellows were killed, and many of those who attacked us were killed. They said I was hit and I did not feel sick."

He told it all, and then he became quiet. When the sun rose he fell down. Something black came out of his mouth. His face became contorted. The people jumped up and cried.

He was dead.