(return to Bartlett index)

Remembering: A Study in Experimental and Social Psychology

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




THE greatest stimulus to social change probably always comes from outside the strict limits of the changing group. This is the primary psychological reason why large and important organisations, such as an army, have groups of attaches, or liaison officers, with alien friendly organisations; why commercial undertakings or other associations often form selected groups among themselves with more intimate interrelations than they possess with similar bodies outside of the special grouping, and why, in the world as we know it, there have grown up any number of elaborate methods of facilitating contact between social groups. Yet when a technique, a custom, or an institution is adopted into one group from another, by whatsoever means, the selective conservation of the recipient group always works it into a pattern which is distinctive of itself. It is this process of the development of characteristic patterns within the group, into which all alien material that is retained must be fitted, that we are now to study. The emphasis passes, for the moment, away from psychology, in the strict sense, towards sociology. It is not with emotions, images, ideas, individual attitudes, that we are concerned, but with objective changes of culture. The general name which, as I have already stated, I propose to use to cover the whole of the processes involved is conventionalisation. The problem is: Here is an element of culture coming into this group from another. What are the main principles of the changes it must undergo before it finally settles down to an accepted form in its new social setting? The range of this question is immense. In a brief discussion of a problem which, adequately treated, would need far more than a lifetime of study, I cannot avoid an appearance of dogmatism.

When cultural material is introduced into a group from the outside it suffers change until it eventually either disappears or reaches a new stable form. The main principles involved in the production of the new specific social form are:

(a) by assimilation to existing cultural forms within the receptive group; [268]

(b) by simplification, or the dropping out of elements peculiar to the group from which the culture comes;

(c) by the retention, in a number of cases, of details peculiar to the communicating group, but apparently not centrally connected with the custom, or product, that is adopted;

(d) by a genuine process of social constructiveness. Of these the only one that is not fairly obvious, and that raises definitely controversial issues, is the last, and this will have to be considered in some detail, when the other processes have been briefly illustrated and discussed.


(a) By Assimilation

Among the North American Indians the Abrahi, and, in particular, the Passamaquoddy division of the tribe in the State of Maine, began, towards the end of the nineteenth century, to engage in civilised industries. In consequence of this, they introduced a method of keeping accounts of business transactions. Already they possessed fairly highly developed ideographic methods of keeping records. These were used for the new purpose, but were supplemented by the imitation of symbols which the Indians took over from the white communities. Here is a typical instance:

This, being interpreted, means that a customer, an old woman, a descendant of an ancient tribal name 'Owl', bought on credit one plug of smoking tobacco of oblong shape, like the usual packages of this material. Also she bought two quarts of kerosene oil. The price was twenty cents. The account was settled by the barter of a basket (1).

All the concrete objects in this receipt form are represented by already conventionalised ideographic signs: the tribal name 'Owl.' by the tree and the bird; the smoking tobacco by the oblong package with [270] smoke ascending; the oil by the two quart tins and a lamp, and the basket by the simple drawing to the left of the form. Among these are introduced the single and double vertical strokes for quantity, the two Roman numeral forms for the twenty cents, and a peculiar cancellation sign to the extreme right to indicate settlement of the account. Many such forms became current in this particular group. Beyond doubt each of such forms was produced by an individual, but by an individual directly determined by an ideographic system that was a social fact before he became a member of the group. The new features, the numerals and the cancellation sign in this instance, were assimilated to existing methods of picture writing and the whole forms a cultural detail peculiar to the group in question.

Again, years after the Spaniards had conquered New Granada, when the native Indians were all accounted Christian, and seemed to have taken over, with little or no modification, the religious paraphernalia of their conquerors, secret Indian shrines were sometimes found. In one of these was discovered, offered to the "overthrown" idols, the cap of a Franciscan friar, a rosary, a priest's biretta and a Spanish book of religious precepts (2).

Almost any popular story cycle yields illustrations of the same process. When a particular story series is current in any social group, if a new theme is introduced from another group, it is nearly certain to get eventually worked into the existing series, and it then acquires a twist in the direction of the special characteristics of this series. For example, into the Raven legend current among the Tsimshian Indians are incorporated any more or less suitable stories that come along. There has been a strong social tendency in this group to work into the Raven myth cycle any tale that will fit into the series of adventures recorded. Some of the stories, particularly in the Tlingit versions, occur independently, so that it is possible to see the changes which they undergo with assimilation (3).

The multiplication of illustrations is unnecessary. There is no process more common than this one of social assimilation to an existing cultural background. It affects every department of public life, and it points to at least one important conclusion. Never is it mere social contact that produces the transference of cultural features from one group to another. Those features only are transferred for which already [271] there exists a suitable background in the social possessions and functions of the recipient group.

The details of the process of transfer by assimilation merit a far more intensive study than has yet been made of them. For example, an examination of available instances will demonstrate beyond doubt that certain features are exceedingly liable to undergo transformation in the course of this process. Among these are proper names, colour significance, and the uses made of concrete objects. When, as in the case of many ritual performances, an order of sequence must be maintained, special devices, turning particularly upon the use of song, rhythm, and muscular reactions, are apt to be utilised. However, the minute details need not detain us now. The general effect of social assimilation is to produce cultural patterns which are distinctive of the social groups concerned.

(b) By Simplification

That most cultural materials received by one group from another undergo considerable simplification of detail in the course of time again goes almost without saying. Here, as in the first case, the results of my experiments have their frequent parallel in social life. What usually happens is that some element of an original complex gradually attains a more and more important position and comes to stand for all the rest. The long story of the development of the common alphabetical forms is a case in point (4). Usually simplification does not take place immediately, but grows gradually in the community, as the material concerned passes through many hands, or the customs introduced become more and more a matter of the everyday life of numerous people. Each particular change probably comes about in a totally unwitting manner. No contributor sees the end towards which he is progressing; and yet, when a whole series is available, the different stages may appear to be progressively connected.

Cases abound in practically every detailed study of the development of decorative art forms (5), or of the evolution of articles of material culture. It is not necessary to present instances in full detail here, and I will content myself with one example, taken from Mallery's study of American Indian "Winter Counts". Most of the tribal groups of [272] North American Indians possessed systems of picture writing. Numerous series of these, called Winter Counts (6), depict outstanding events of various hunting or war periods. Here is a brief selected history of a pictographic sign for 'raising a war party' among the Dakota Indians:

The first of these reproductions is a complete pictograph which shows the war eagle, the Indians coming together, the calumet, or treaty pipe, the scalp and the tomahawk. In the second, the calumet remains in an abbreviated form and there is a single Indian with tribal marks; a single feather replaces the war-eagle. The third case is nothing but a conventional symbol, with remains of the calumet and of the single feather; it is a form interpretable only within the special group, or by someone who has made a study of this group; and it is new, but has been reached through a process of progressive simplification.

Not infrequently simplification is obtained as a final stage of a process which first displays a great amount of elaboration, the elaboration being secured commonly by reduplication and repetition of detail. So common is this in decorative art that Haddon speaks of it as a characteristic device of the decorative mind. In Papuan ornamentation curved lines, reduplicating the angle of a mouth, as in fish designs, are often found. "In some cases, as in undoubted fish, they are certainly intended for gill-slits: in others-for example in crocodiles-whatever they are intended for they certainly are not gill-slits. In the latter cases, they may be merely the expression of that tendency to reduplicate the motive or design which appears to be characteristic of the decorative mind; on the other hand, it may be an example of a transference of features or attributes which often occurs in the art of savages" (7). Some very interesting instances both of elaboration and of [273] simplification may be drawn from a study of the development of the African throwing knife (8).

Here, as in the case of transfer by assimilation, there is need for a far more detailed sociological study of the principles and devices involved than has hitherto been attempted.

(c) By the Retention of Apparently Unimportant Elements

In commenting upon the results of some of my experiments, particularly upon those obtained from Serial Reproduction, I have already discussed the curious tendency to preserve the apparently odd, trivial, disconnected and novel element. This occurs very often in conventionalisation. It rightly forms a strong point in the evidence put forward for ethnological development through the contact of peoples, as this has been marshaled by Prof. Elliot Smith and W. J. Perry. An apparently trivial likeness in unimportant detail may be a strong link in a chain of argument intended to demonstrate cultural contact of different social and racial groups (9). Most of the available illustrations require a considerable amount of explanatory description, but one or two may be given which can be treated briefly.

Dr Haddon shows that all the decorated canoes from a certain part of British New Guinea possessed "a peculiar incised decoration a short distance from the bow, usually near the termination of the weatherboard. This received the same name, Koimai, as the cicatrice formerly carved on the shoulders of the men " (10). By comparison of instances, it became clear that this decoration was originally a full realistic reproduction of a man's face. The representation passed through many stages of simplification, became a mere conventional pattern of concentric triangles and finally, only the single incised design, which was, to begin with, an unimportant bit of the whole complex decoration, was left. This remnant was so built into the canoe design as to become a perfectly characteristic mark of vessels coming from a definite area.

Similarly, in a great many instances of decorated arrows, also from New Guinea, an original crocodile figure undergoes striking changes. The mouth drops out, the eyes are transformed, the fore limbs disappear; but two features, which would at first seem less prominent than the others, are retained and often exaggerated. These are: "the remarkable retention of the projecting nostril, which may often be [274] found as a slight prominence in very degraded arrows", and "the still greater persistence of the tail and hindquarters of the crocodile". The latter instance, however, Haddon suspects to be "due to the striking decorative effect of the concentrically marked cloacal plate" (11).

In many instances, the retention of an apparently unimportant feature, together with the simplification or decay of other detail, may give an entirely new bias to the material concerned. Thus the snake arrow may evolve from the crocodile arrow by the preponderant influence of a persistent nostril; or, from the same beginning, the lizard arrow may develop, as a result of the persistence of tail characters (12).

Perhaps, although this is more than a little doubtful, the occurrence in Chinese bird chariots of a loop on the chest of the large bird, and of an opening in its back with zoomorphic decoration, are instances of the persistence of merely accessory bird detail in the original design (13).

Although the persistence of the 'trivial' can be dealt with merely as an objective fact characterising many processes of conventional change, it does at the same time raise some intriguing psychological problems. After all, these detached small features of a complex pattern which resist change for prolonged periods must somehow have a strong significance for those who retain them. It is possible that this significance may, in some cases, be no more than that, standing apart from those features of the pattern which have an immediate social background, they are most of all free from the transforming influences of assimilation. But in other cases, the significance may be of a more positive order; and the apparently trivial may persist because, in fact, it has a hidden significance of a symbolic character which sets moving some deep-seated social or individual tendency. It is useless merely to generalise. Each case has to be taken on its own merits. That small features often do resist change in an astonishing manner is demonstrable, and it is one of the facts which go far to justify the view that the similarities of culture which may fairly be used as arguments for social contact are generally similarities in relatively detached, unimportant and odd details.

(d) By Social Constructiveness

The three processes just briefly discussed, together, no doubt, with a number of others, which a full examination of conventionalisation would have to include, can all be treated, if it is so desired, as the [274] direct outcome of social institutions, and independently of a study of the attitudes of the individuals belonging to the groups concerned. These processes, when combined, give rise to cultural patterns which in many cases are found nowhere else than in the receptive group in question, and are thus uniquely characteristic of this group. In the sense that they build up something new, they may all fairly be described as constructive. Yet they do not, I think, quite give the full picture of social constructiveness.

The main reason for this is that every well-established social group possesses not only a structure which has been built up in its past, but also a function, or a group of functions, within the community of which it is a part. These functions have to be expressed in co-ordinated human activity, and all such activity has not only a history, but also a prospect. We can say of it that it inevitably tends to develop in certain more or less specific directions; and, if we know enough, we can state in some detail the paths along which it is tending. Since, then, a group is maintained by its activity, as well as by its more or less permanent structure, it is possible to say that the social group, as such, possesses a certain trend of development. This trend need not be, and in the majority of cases it certainly is not, present in the mind, or fully represented in the behaviour, of any individual member of the group. Nevertheless, it is a genuine social factor which may determine social change within the group. So, when any cultural features come from outside, they may be transformed, not only by assimilation, by simplification and elaboration, and by the retention of apparently unimportant elements, but positively in the direction along which the group happens to be developing at the time at which these features are introduced. Since this direction need not be fully exhibited --to say nothing about its being fully formulated-- by any member of the group, its effect in giving a positive bias to change in a certain direction may fairly be called social. The most general effect of this positive influence is probably to weld together elements of culture coming from diverse sources and having historically, perhaps, very diverse significance: hence it is definitely of the constructive order.

Such welding is a characteristic reaction towards imported elements of culture adopted inevitably by all strong and vigorous groups. It means that the imported elements change, both in the direction of existing culture and along the general line of development of the receptive group. It also means that while probably the most effective stimulus to change comes in the main from social contacts, important [276] social forms of culture may genuinely grow up within the group. The main way in which this social constructiveness is exercised is that all incoming cultural elements, whatever their origin, which are related to the same general sphere of life, tend to be dealt with by the influences which determine the trend of development of the receptive group in relation to the sphere of life involved. Consequently the imported elements will be built together and worked up into forms, the details of which come from varied sources.

Perhaps it ought to follow from this that as the basis of group organisation becomes more and more definitely formulated, and as the range and diversity of a group increase, social constructiveness becomes more marked and more frequent. I incline to the view that this does, in fact, happen. But the case is greatly complicated by the fact that probably the individual, on the whole, acquires greater influence in the community as the complexity and range of social grouping grow. Thus, besides that constructiveness which marks a true social function, we have to allow for the fact that individuals of a dominantly constructive turn of mind tend to play a larger and larger part in the mechanism of its expression, as society develops towards the phase of modern civilisation.

A few instances of social constructiveness must now be examined, though no attempt will be made to work them out in full detail.

Modern developments of science, both in practical and in theoretical directions, demand a great amount of team work, in which specialists belonging to different fields must co-operate. How, in all such cases, the final product of the team is achieved presents an interesting series of problems.

For example, during the War of 1914-1918, the demands of aircraft defense stimulated every large nation group concerned to develop mechanical, or semi-mechanical, devices for the detection of attacking aircraft at night. These all, by physical and physiological necessity, followed broadly the same lines; but there were important differences which, for the time being, were not fully known as between group and opposing group. Since the end of the War, all the important national groups in Europe, together with America and some other nations, have pursued the matter further. To some extent these groups have been in contact one with another. In every case there have been many developments, so that the apparatus coming into use now possesses many points of difference when compared with that employed during the late War. No instrument can be said to be the work [277] of any single man, but of a number of men. In most cases, several features of the present instruments have been transferred bodily from one group to another. Not only is no complete instrument the result of the foresight of any one person working alone, but it is not simply the aggregation of the contributions of a number of different men, all belonging to the same army unit, or to related units. A, perhaps, proposed this; B that; C the other thing; and E, very likely proposing no specific detail himself, worked all the details derived from tie various sources into a practical form, so that the A, B and C details are not any longer exactly as A, B and C thought of them. More than that: when the apparatus came into experimental use, it suffered various modifications of its functional parts which nobody ever thought out very clearly, if at all. Some of these are particularly interesting. The commonest aircraft detection instruments have to be controlled by a group, or by a team. There are some forms which demand very much greater interdependence among the members of the team than is the case with others. Each has developed within its own special social milieu, so that a well-instructed onlooker, asked to furnish a rationale for differences in the type of instrument in common use; will often find himself speaking in social, group terms. Yet it is fairly certain that nobody ever put this sort of characteristic before himself as an ideal when he was thinking about the instrument. It simply worked out so in practice. Perhaps, in all team work of this order, when the construction of instruments, or of theories is concerned, the details, and even the final form, come from some individual or other, and must have been to some extent articulated or planned. But the group trend is apt to come in by way of unwitting modifications produced by practice.

It is interesting to compare this team construction of an instrument, a theory, or a plan of administration, with the achievement by a group of a fluid, free, yet fairly stable social organisation, like that say of a Rugby XV, or of a cricket eleven. Here is a group with a temporarily fixed leader, but in actual fact any man may be pivotal at a given time in a given game; and, especially in a game like Rugby football, the rest slip easily into positions about him. Nine-tenths of a swift game is as far as possible from the exploitation of a definite, thought-out plan, hatched beforehand, and carried out exactly as was intended. The members of the team go rapidly into positions which they did not foresee, plan, or even immediately envisage, any more than the bits of glass in a kaleidoscope think out their relative positions in the [278] patterns which they combine to make. Yet a team has its characteristic persistent 'style' which determines the players all the time. In England, the Harlequin style in Rugby football is not that of Blackheath, or the Army style that of the Navy. At one time it became very nearly a tradition to compare the play of the Scottish forwards with that of the Welsh three-quarters, to contrast Irish turbulence with English persistence; and all of them are different from French individuality. Every now and then, in a skilful, hard game, it is possible to see a new team organisation flashing out, built upon group qualities and the swift practical insight of one or two individuals, but not thought out or foreseen by anybody as regards most of its details. In cricket, also, the same characteristic group achievements often force themselves upon the attention of the critic. "A Lancashire and Yorkshire match is inconceivable at Tonbridge. In the north, even a Saturday afternoon league match is grim enough; a Lancashire and Yorkshire match in little. The game as it is played in the Lancashire leagues at the week-end is never a laughing matter, never a summer-time amenity to be watched from deck-chairs with a tea-cup in hand. (They have an interval for tea between innings in the Lancashire league, true, but usually it is as serious as the cricket; a high tea with meat to keep the flesh strong and the temper eager.)" (14)

Perhaps, in more serious affairs still, a group achievement of an oligarchy, or a democracy, or even of a despotism, or of practical, constitutional devices grows up in similar ways. We can put our finger upon this, that, or the other thing and say: "This comes from such and such an individual source. But when we have done all that can be done in this way, there is much left over. It is left, not merely because the phenomena are too complicated, but because any constructive achievement of social organisation depends upon the form and trend of the group before the achievement is effected, as well as upon the efforts of innumerable individuals in the mass.

A small enough group, a Rugby XV, or a small committee, for instance, can sometimes be observed which has temporarily lost, or never has achieved, this team organisation. An individual has to stop and think what another is going to do or say before he can play his part, and then it is too late. There may be as much individual constructiveness as ever, but something blocks the weaving of the group pattern, and the result has a disconnected appearance, as if no real constructiveness were present at all. [279]

In his very interesting study of Propaganda Technique in the Great War (15), G. H. Lasswell shows how each person responsible for public propaganda must adapt himself "to territorial prejudices, to certain objective facts of international life and to the general tension level of the community". By the last of these factors, which is especially interesting to us now, is meant the present state of adaptation of a social group and the cumulative social effect of its past history. "By the tension level is meant that condition of adaptation or of maladaptation, which is variously described as public anxiety, nervousness, irritability, unrest, discontent or strain. The propagandist who deals with a community when its tension level is high, finds that a reservoir of explosive energy can be touched off by the same small match which would normally ignite a bonfire". No study merely of the individual member of the group will suffice to gauge this public level of tension, and yet in accordance with it the general direction of propaganda during the Great War was inevitably, though not by 'malice aforethought' different in each national group. Germany relied greatly upon the sympathy of her old nationals and on pride in her own achievements; France largely upon hatred of ancient foes and the use of important historical words like 'humanity' and 'democracy', etc., which possessed a special significance for themselves "and reverberated with a tremendous clang abroad"; Great Britain upon humanitarian appeals and the diplomatic discouragement of the enemy allied forces. The actual social mechanisms set up to control and develop propaganda, again, showed characteristic persistent differences in one nation group as compared with the others.

Any sudden fury of effort arising within a group, such as the Elizabethan outburst in literature, discovery and colonisation, where, though in different fields of culture, the same characteristics recur, seem to force us towards the notion of a socially determined constructiveness. The study of the building up of a religious or a political group, such as the rise of the Society of Friends among the ferment of Puritanism, or the early growth of the English Labour Party, give illustrations of the achieving of complex social structures which, at point after point, are determined by social trends. One person contributes this and another that, but so far as their work stands their contributions, except in very unusual circumstances, must all be in the same trend. The social drive limits and directs them as it does the Dahomey artist: "In vain does the Dahomey artist convince himself [280] that he is following a new design in the execution of which he is merely following momentary promptings. Though he believes himself to start and to continue without any conception of the figure which he is about to produce, an examination of the procedure of such a native artist reveals the existence of 'determinants'. When 'turning' in his freehand design, he must not make smooth curves, but put a characteristic 'kink' in each. Moreover, having started his design, the rest of the figure must fall into a certain harmony of outline and balance of parts which, of course, limit individual choice. These characteristics are imposed by the culture, the artist merely varying the prescribed form, though never departing from the general rules laid down by the conventions of the group. We find the creation of new objective designs, but the newness lies within well-defined social limits" (16).

Yet, however much agreement there may be as to the fact of social constructiveness, we know almost nothing as to its exact mechanism -the lines of its most ready expression within the particular group, the limits of its achievement, its exact relation to individual effort. These all constitute important sociological and psychological problems which will provide a great field for future research.


Conventionalisation is a process by which cultural materials coming into a group from outside are gradually worked into a pattern of a relatively stable kind distinctive of that group. The new material is assimilated to the persistent past of the group to which it comes. It is simplified in certain directions, perhaps elaborated in others; often it retains oddly unimportant looking foreign elements; and it is moulded into a characteristic complex form by many influences, among which is something that we must call a social trend. Nothing that I have said so far would indicate that the social past which inevitably helps to shape a group's new acquisitions persists in any other way than in its institutions, its current traditions, and its preferred persistent tendencies. Others have thought that there may be something more than this. Here it is that we come most directly upon the question of a 'collective unconscious'. It is a tangled subject, but to its study we now must turn.


(1) See Mallery, "Picture Writing of the American Indians", Ann. Rep. Bur. of Amer. Ethnology, 1888-89, p. 201.

(2) R. B. Cunningham Grahame, The Conquest of New Granada, 1922, p.

(3) Franz Boas, "Tsimshian Mythology", Rep. Ann. Bur. Amer. Eth. xxxi, 571 ff.

(4) See I. Taylor, The Alphabet, London 1883.

(5) See e.g. Haddon, "Decorative Art of British New Guinea", Cunningriam Memoir8, No. x; C. H. Read, Journ. Anthrop. In8t. xxi, 139-54.

(6) Mallery, op. cit. p. 652

(7) Haddon, op. cit. p. 46.

(8) See E. L. Thomas, Journ. Roy. Anthrop. Instit. Lv, 129.

(9) See, e.g., Elliot Smith, The Evolution of the Dragon, Ch. II, Manchester 1919.

(10) Haddon, op. cit. p. 59.

(11) Haddon, op. cit., p. 53; figs. on p. 55.

(12) Idem, p. 58. C. G.

13) Seligman, Journ. Roy. Anthrop. Instit. LVIII, 247.

(14) Neville Cardus, Cricket, London 1930, pp. 173-4.

(15) London 1929.

(16) W. D. Wallas, "Individual Initiative and Social Control , Amer. Anthropol. N.S. xvu, 1915, pp. 647-68.

Copyright note: Posted with permission of Cambridge University Press.