Skip to main content

Full text of "Discussions on child development"

See other formats



Digitized by the Internet Archive 

in 2010 with funding from 

Lyrasis IVIembers and Sloan Foundation 




Child Development 

A Consideration of the Biological, Psychological, and 

Cultural Approaches to the Understanding 

of Human Development and Behaviour 



M.D., PH.D., D.P.M. 

Lecturer, Institute of Child Health, University of London 


Professor of Child Psychology, University of Geneva 


The Proceedings of the First Meeting of the 

World Health Organization Study Group 

on the Psychobiological Development of the Child 

Geneva 1953 

New York New York 


IS6 r 7 

Aj qfsi 7^ 


V / 

c ' :l 



DR. JOHN BOWLBY Psychoanalysis 

Director, Children's Department 
Tavistock Clinic, London 



Josiah Macy, Jr. Foundation, New York 

Research Promotion 

DR. G. R. HARGREAVES Psychiatry 

Formerly Chief, Mental Health Section 

World Health Organization 

Professor of Psychiatry, Leeds University 


Professeur de Psychologic de I'Enfant 

Institut des Sciences de I'Education de 

rUniversite de Geneve 


Forschungsstelle fiir 

Verhaltensphysiologie des Max-Planck 

Institutes fiir Meeresbiologie 

Buldern iiber Dulmen, West Germany 



DR. MARGARET MEAD Cultural Anthropology 

Associate Director Dept. of 


American Museum of Natural History, 

New York 


Director, Clinic for Convulsive Disorders, 
Stora Skondal, Stockholm 


Charge de Cours de Neurophysiologie 
appliquee Universite de Geneve 


Professeur de Psychologic a la Sorbonne 
et h r Universite de Geneve 


Charge de Recherches, Centre National 
de la Recherche Scientifique, Paris 



Formerly, Associate Director 
Rockefeller Foundation, Paris 

Research Promotion 


Formerly Senior Lecturer, 

Sherrington School of Physiology, 

St. Thomas's Hospital 

Lecturer, Institute of Child Health 

University of London 

Human Biology 


Director of Research 
Burden Neurological Institute, Bristol 



Directeur du Laboratoire de Psycho- 

biologie de I'Enfant 

Institut des Hautes Etudes, Paris 



DR. J. C. CAROTHERS Psychiatry 

Psychiatrist, St. James' Hospital 
Portsmouth, England 

DR. E. E. KRAPF Psychiatry 

Associate Professor of Psychiatry 
University of Buenos Aires 

DR. CHARLES ODIER Psychoanalysis 

Chateau de Vernand, Lausanne 


This volume, and others which will follow it, give an account of an activity 
of the World Health Organization — the Research Study Group on the 
Psychobiological Development of the Child. 

The popular view of the World Health Organization associates it chiefly 
with practical activities concerned with the application of existing know- 
ledge through public health services and as far as the great majority of the 
Organization's activities have been concerned this view is accurate. More 
than half the Organization's funds have from its creation in 1948 been 
devoted to such programmes as the control of communicable diseases. 

But the Organization differs from the International Health bodies which 
preceded it in having placed upon it a specific obligation to 'foster activities 
in the field of mental health, especially those aff"ecting the harmony of 
human relations'. In this field the mass programmes appropriate to such 
problems as the control of malaria are not applicable. In the mental health 
field there are no equivalents to D.D.T. and penicillin, and aetiological 
knowledge is scanty. 

Although the Organization has devoted most of its energies, and its 
funds, to the preventive application of knowledge gained by research in 
national institutions, it is also obligated by its constitution to promote and 
conduct research and it was as a contribution to the promotion of research 
that the Research Study Group was formed. It was not an isolated activity 
but was part of one of the natural trends of development of the mental 
health programme which began in 1949. 

The World Health Organization makes considerable use of Expert 
Committees (small groups of distinguished workers in a given field drawn 
from different countries) to advise it on the development of its technical 
policy. Such an Expert Committee on Mental Health was convened in 
1949. This Committee recommended that it was desirable that the Mental 
Health Programme should concentrate especially on the psychiatry of 
childhood. It emphasized the fact that others apart from psychiatrists 
must be called on to contribute in this field and mentioned specifically the 
anthropologist, the sociologist, and the social and developmental psycho- 
logist. In its report the Committee urged that W.H.O. should also 'actively 
encourage research which sets out to fill gaps in fundamental knowledge' 
in the mental health field, and in this connection it specifically mentioned 
'Research into the biological, psychological, and cultural determinants of 
personality structure'. It is evident, however, that since the funds available 
for the mental health programme have been small — less than two per cent 
of the Organization's budget — the Organization could not itself finance 
research on any appreciable scale. The Organization's contribution has, 
therefore, been directed toward the co-ordinating, surveying, and stimu- 
lating of research. 

Two such W.H.O. surveys on different aspects of the Mental Health of 


Childhood were widely read. The first of these was "The Psychiatric 
Aspects of Juvenile Delinquency' by the late Dr. Lucien Bovet, and the 
second by Dr. John Bowlby on 'Maternal Care and Mental Health' 
surveyed existing research on the psychological eflfects of the continued 
separation of infants from their mothers, or mother substitutes. 

Both these surveys posed questions which gave rise to the thought of a 
multiprofessional discussion group on child development problems. The 
thought was crystallized into a concrete plan by the stimulus of the Oxford 
Conference of the Mental Health Research Fund on 'Prospects in Psychia- 
tric Research' held in 1952, and the model on which the plan was based 
was the informal conferences of the Josiah Macy Junior Foundation. 

The aim of the Group was to bring together once a year for four or five 
years, during a period of a week, a small number of internationally 
eminent workers in the different disciplines which study different aspects 
of the psychobiological development of the child. 

The aim of the meeting was not the reading of papers, the passing of 
resolutions, or the issuing of a report, but the provision of an opportunity 
for mutual understanding to develop between workers in different disci- 
plines, and on the basis of that understanding the attempt to relate the 
findings of one discipline to those of another and the hope that new 
research, and particularly joint research, might be undertaken. The 
Organization was fortunate in obtaining, as chairman of the Group, 
Dr. Frank Fremont-Smith, of the Josiah Macy Junior Foundation. His 
personal qualities, and his great experience of conferences of this type, 
created in this international group (despite the problems of simultaneous 
interpretation) the atmosphere and enthusiasm which made the success of 
the group possible. 

It was not originally intended that the proceedings of the meetings should 
be published; but the mimeographed transcript which was produced for 
the benefit of the members evoked such interest in those outside the group 
who read it that many requests for copies began to be received. Hence the 
decision to pubfish this series. To make the publication possible con- 
siderable editing has been necessary since the original transcript had to be 
reduced by about two-thirds. The Group is much indebted to one of its 
members. Dr. J. M. Tanner, for carrying out this task so skilfully. 

Finally, as the Chief of the Mental Health Section of the World Health 
Organization at the time these meetings were convened, I should like to 
express my personal appreciation to Dr. Brock Chisholm and Dr. Marcolino 
Candau, the first two Directors General of the Organization, and to Dr. 
Norman Begg, Regional Director for Europe, for their support of a venture 
which, although not connected with the day-to-day practice of current 
public health work, may yet through some of its many repercussions have 
its significance for public health workers of future generations. 


Lately Chief, Mental Health Sec- 
Leeds University tion. World Health Organization 


PREFACE page 9 


1 Physical and Physiological Aspects of Child 
Development 36 

2 The Behaviour of New-Born AnencephaUcs with 
Various Degrees of Anencephaly 62 

3 Criteria of the Stages of Mental Development 75 

4 Comparative Behaviourology 108 

5 Electroencephalographic Development of Children 132 

6 Stages of Psychological Development of the Child 161 

7 Psychoanalytic Instinct Theory 182 

8 The Cross-Cultural Approach to Child Development 
Problems 200 

INDEX 237 


FREMONT-SMITH (Chairman): 

Je suis tout a fait content d'etre ici avec vous, mais malheureuse- 
ment je ne peux pas parler le frangais, and so I will go on, not 
speaking in good English, but in the patois of the U.S.A. What I 
now have to say is intended to be introductory, and to tell you what 
we would like to have as the mood of the conference. I have been 
co-opted to this job because it was hoped that we could use at least 
some of the Macy Foundation conference methods in this meeting, 
and so first I should say a little about them. The Macy Foundation 
is a charitable body that makes grants for research. In the course of 
doing this the Foundation directors and officers became disturbed by 
the narrowness of the approach of the investigators who asked us for 
help. Their projects were drawn up from a unilateral point of view, 
and it seemed more and more that, to make advances in practically 
any problem in science, one needed the participation of several dis- 
ciplines. We found that the investigator was frequently ignorant of 
the contribution already made or potentially to be made by another 
discipline, and too often was uninterested. The general position that 
we held was that nature is all of one piece and that the departments 
of universities, the speciahties which had grown up for good and 
necessary reasons, tended artificially to divide that nature and to set 
up barriers which prevented communication between the different 
disciplines. Our conference programme arose as an eflTort to rein- 
tegrate the scientist's approach to nature's secrets, to try to bring a 
genuine multi-disciplinary point of view to bear on problems. How- 
ever, we discovered that this was not as simple as it might seem. We 
tried at first to crowd the meeting with speakers representing different 
fields, as in the usual kind of meeting. It didn't get us very far. We 
tried to bring together the two or three men from different depart- 
ments in one university who were working on the same problem. 
They didn't hke it. Often they didn't even like each other. We then 
gradually began to focus more and more on what are the obstructions 


between disciplines and now we look upon our conferences as experi- 
ments in communication and as efforts to identify and remove, if 
possible, some of these obstructions. 

One thing that seemed to come out of this thinking was that there 
is an over-emphasis, derived, I think, primarily from our universities, 
on the intellectual side of learning and the intellectual side of science 
— and I mean intellect as opposed to emotion, the logical as opposed 
to the creative — and we felt that we would like to dp something to 
redress the balance. We felt that the lecture system of the universities 
had reached a point where it was practically limited to the making of 
statements at people rather than being concerned with communica- 
tion with people. This applies also to the papers at the ordinary 
scientific meetings, which are usually a series of statements made at 
people, statements which give great satisfaction to the speaker but 
little to the listener. Since the scientific programme is crowded, only 
one or two questions may be raised at the end of each statement, and 
then one goes on with the next paper. We wondered if something 
might not be accomplished if we reversed the process and tried to 
make the questions and answers, the discussion, the heart of our con- 
ference, and the papers as small a factor as possible. Gradually we 
pushed the papers back and brought the discussion forward so that 
now we have somebody who makes introductory remarks for about 
twenty minutes, which serve as the basis of discussion for the rest of 
the day. We tried to bring the discussions that ordinarily take place 
in the corridor outside the scientific meetings into the meetings 

■ With respect to making statements at people rather than communi- 
cating with them, it seems to me that we should pay more attention 
to the hstener. We should ask ourselves what kind of a receiving set 
the other fellow has got. Some of the difficulties of communication 
are linguistic; but others are to do with the imprint of authority, 
which makes it impossible for us really to hear a viewpoint which 
challenges the authority with which we have identified ourselves — 
sometimes it is our own authority. In days gone by I think it was 
possible for a person to remain relatively static for long periods of 
time, but today I think we must expect change. But change is some- 
thing which produces anxiety, and anxiety is apt to be transformed 
into hostility. And I think that is one of the reasons why, when 
someone is challenged with an idea which would produce a change in 
his approach to the problem, so often the response is a hostile one. 
These resistances, these obstructions to communication, these dis- 
torting lenses, may change with diff'erent environmental situations. 
When the environment is a chill and anxious environment, the lenses 
go on — it's like the guard the boxer puts up. But if there's a warm 


and friendly atmosphere, then the glasses come oflf, the guard is 
lowered, and the way is open for communication. That's one of the 
reasons why people must have an opportunity to be associated with 
each other informally. It's one of the reasons why having the same 
group, with a few additions, meet again and again and again often 
helps so singularly well to provide a basis for communication. 

We try to encourage the participants in our meetings to speak 
freely, not to hesitate to ask a fooUsh question, because how can 
they be sure it won't evoke wisdom in someone else ? We are not here 
in this kind of group to solve a point, to defend a view, but rather to 
examine our own bhnd spots, or at least to give others a chance to 
point them out to us. In this way we hope that we can open the doors 
to the kind of communication that previously hasn't been possible. 
This kind of meeting should be primarily for the cross-fertilization of 
ideas, for the meeting of minds, for the stimulation of curiosity, and 
for examining and bringing out the resources of the group. The 
resources of each of us are certainly not known to the others, and 
some of them possibly not to ourselves, because we have all stored 
many memories and experiences which we cannot bring to the 
surface at will on a given problem, but which may be there in the 
background of our memories, and brought out by someone else's 

We will now introduce ourselves to each other. I will start with 
myself and then ask each of you to say in a few words how you got 
the kind of interest that brought you to this meeting. I had an 
Hungarian mother, which has helped to give me a warm feeling 
towards the Continent, and a father of New England origin, and I'd 
like you to feel that when I am quite wild and woolly it's the romantic, 
dynamic, Magyar spirit bursting out and when I'm sensible and calm 
and intelligent then it's the New England spirit, which, of course, 
came from Great Britain. I graduated from the Harvard Medical 
School in 1921 and had a training in pathology and then another 
year and a half in medicine, and then went to the Department of 
Neuropathology in the Harvard Medical School under Dr. Stanley 
Cobb and did research on cerebro-spinal fluid, and on body fluids in 
general: respectable, solid, basic research. Then, when I was 
measuring cerebro-spinal fluid in a manometer, I discovered that 
even such a simple measurement as that could not be made accurately 
unless I had some knowledge of the emotional state of the patient, 
because if the patient was tense he might either hold his breath and 
give himself a high venous pressure, and too high a pressure in the 
cerebro-spinal fluid, or, on the other hand, he might react by over- 
ventilation, lower his venous pressure and get too low a cerebro- 
spinal fluid pressure ; in this way I was introduced into psychosomatic 


problems. Then, since Dr. Cobb had many patients with epilepsy, I 
became very much interested in this problem and particularly in the 
factors which precipitated convulsions, and especially in emotional 
factors as one of the precipitating agents. And from there it turned 
out that the emotional factors which precipitated such attacks were 
very often factors of which the patient was unaware : old conflicts, 
for which the patient had an amnesia, so that I became more and 
more drawn from the psychosomatic into the psychiatric field, and 
although never properly trained in psychiatry nor properly analysed, 
I did have a personal analysis, which was very good for me, and some 
psychiatric training. Then I came into the Macy Foundation in 1936 
and since then I have been in a curious situation. You all know the 
definition of the specialist who gradually gets to know more and more 
about less and less until he knows practically everything about 
nothing, and the generalist who gets to know less and less about 
more and more until he knows practically nothing about anything; 
but I am a mixture of those two, because I get exposed to an enormous 
range of conferences where the participants are groups of highly 
trained specialists. 


After medical studies in Geneva, Zurich, and Vienna, I worked in 
the Physiological Institute of Zurich, where I wrote my doctor's 
thesis under the direction of Professor W. R. Hess. I was already 
interested at that time in the correlations between psychological 
functions and the vegetative nervous system. From 1931 to 1934 I 
got a training in psychiatry as assistant at the Psychiatrische Univer- 
sitatsklinik of Zurich, and from 1934 to 1937 a training in neurology 
at the Chnique des Maladies Nerveuses at the Salpetriere in Paris. I 
went then as a Rockefeller Fellow to the Neurological Institute of 
Northwestern University in Chicago, where I studied the functions 
of the reticular system of the brainstem under the direction of 
Professor S. W. Ranson. During 1938, I worked as assistant in the 
Service de Neurologic de I'Hopital Cantonal de Geneve and in 
1939 I became Chef des Travaux at the Institut de Physiologic de 

In 1941 I went to Zurich as Chef des Travaux at the Physiological 
Institute, where I worked again with Professor W. R. Hess. I was 
interested in the integrative activity of the nervous system and had 
the opportunity of studying, with Professor H. Willi, head of the 
Kantonales Sauglingsheim, the development of motor functions in 
normal and anencephahc newborns. In 1948, I returned to Geneva 
to start the Laboratoire de Neurophysiologie appUquee, which had 


as its purpose the bridging of the gap between experimental and 
clinical neurophysiology. Here I had the opportunity to develop with 
Professor Piaget and Mile Inhelder a co-operative study of the 
correlation between mental development as analysed by various 
psychological tests, and the development of the electrical activity of 
the brain, analysed by means of electroencephalography. This is 
one of the reasons why I am here. 


I don't know of anyone who is in this group who has less right to 
be here than I have as I am not trained as a scientist in any particular 
field. I had some experience in the army in the first war as a physician 
and also in general practice in the country and I eventually entered 
paediatrics, in the days when most of paediatric work was a question 
of whether you had sufficient lactic acid in the baby's feed. After 
twenty years in general paediatric practice in Montreal, during which 
I did some work on childhood tuberculosis as a social problem, and 
on the definition of activity of rheumatic fever in children, I found 
that my main interest was in undergraduate medical education. 
Because of that interest I am now dedicated to work in the Rocke- 
feller Foundation as Director for Medical Programmes in Europe. 
The only justification for my attendance here is that I can qualify 
under Dr. Fremont-Smith's definition of one of those who knows less 
and less about more and more, and I think I have almost reached the 
apogee of knowing nothing. 


I am a human biologist, with particular interests in physiology, 
child development, and genetics. I had originally a mathematical 
training at school, with the intention of becoming an army engineer. 
However, this project had never really appealed to me, and a few 
days before it came to fruition I deserted and began a pre-medical 
course. This I greatly enjoyed and soon I had a particular interest in 
genetics and evolutionary theory. Early in the war I was a medical 
student and the Rockefeller Foundation generously offered to take 
some of us from Britain over to America and to train us there because 
of the difficulties we were facing in the hospitals in London. I had at 
that time done a small amount of research work in pure physiology 
and my interests by this time had become, if not focused, at least fairly 
concrete, so that when I was asked about them at the interview 
associated with this trip, I said 'I want to work in the place where 
physiology, psychology, and sociology meet'. This is a stand from 
which I have never really departed. So, having had my anatomical 


and physiological training in England I had my chnical training at 
the University of Pennsylvania, and later on as a house physician 
on the medical service of the Johns Hopkins Hospital. 

Following that, I came back again to England and being now 
quite clear that I wanted to become, if necessary all by myself, a 
human biologist, I spent the next two years in psychiatry, deahng 
almost exclusively with the superficial psychotherapy of neuroses of 
a combat type. I was particularly concerned at that time with group 
therapy. At the end of this period, having learned just enough about 
psychology and psychiatry to understand the language, I went to 
Oxford University as lecturer in Physical Anthropology. I did that 
because it seemed to me that if one was concerned with studying 
human biology and behaviour one should build one's science from 
the ground up, studying first of all the simplest thing there is to 
study in the human, which is what he looks hke. 

At Oxford I spent a great deal of time getting to know the litera- 
ture on the physical growth of the child, and trying to define and 
classify differences in physique among adults. I think the interest of 
these physical differences hes in the light they may throw on differences 
in physiological function and in behaviour. After I had spent three 
years at Oxford learning, by the well-known mechanism of teaching, 
the subject of physical anthropology, I went back to be a professional 
physiologist at St. Thomas's Hospital in London University, where I 
now teach physiology and do research work on the physiological 
differences between people, their genetical basis and their anatomical 

There is one other thing I should tell you. I feel very hkely that my 
best function here may be more to answer questions that the etholo- 
gists and the psychologists may bring up than to propose anything 
of my own, and therefore the more I tell you about the fields in which 
you may reasonably expect to get a sensible answer out of me the 
better. For the last four years I have been closely associated with a 
study on human growth run and financed by the Ministry of Health 
outside London and here we have a longitudinal growth-study, the 
first of its kind in England, modelled to a considerable extent on 
those which are doing such good work in the United States. We have 
there about 250 children and a group of us, collected mainly from 
various departments of London University, descend on these children 
two consecutive days every month. We therefore see every child 
within two weeks of its birthday or half-birthday; we see every child 
every six months, or every three months during puberty. At the 
present time we do a number of physical measurements of the children, 
we take photographs in a rather highly standardized and particular 
way which enables us to measure them, and we take a considerable 


number of X-rays, the function of which is to differentiate bone and 
muscle and fat so as to see the growth of the different sorts of tissue 
in children. We have dental people who take X-rays of the jaws and 
are concerned with the development of the teeth ; and a paediatrician 
who does the chnical examinations. We are dealing with bone-ages, 
teeth-development-ages and such-hke things, subjects in which you 
may be interested. We do not have — and I regard it as a very grave 
gap in our investigation ; it is a matter of money and space as usual — 
anybody who is studying the physiological and biochemical develop- 
ment of the children, so I have no personal experience of that. How- 
ever, I am quite well acquainted with such literature as there is on it. 
Also we do not have at the present time any psychiatric or psycho- 
logical studies in progress, and this is a field about which I wish to be 
informed, and know, practically speaking, nothing. 


I started my training as an anthropologist at the age of about three, 
which gives me some sort of qualification for working on the problem 
of children. My mother was a sociologist who was working on the 
adjustment of Italian immigrants in the United States, and my 
grandmother was very much interested in child thought and imagina- 
tion. She was a teacher of young children. So I was trained on my 
younger sisters to record children's behaviour, and by the time I was 
nine or ten I was a fairly competent recorder. We don't know yet 
whether starting that early does any good or not, but I went on a 
fairly straight line, having come out of this sort of academic expecta- 
tion. I took my M.A. in psychology and worked initially on the 
effect of language spoken in the home of immigrants on their perfor- 
mance in intelligence tests at school, and that meant working through 
the literature on the relationship of intelligence testing to race. I then 
went into anthropology under Franz Boas. I suppose I belong 
ethnically to that very small and rapidly vanishing minority calle'd 
'Old Americans', as I have no ties that can be traced to Europe. 1 am 
tenth-generation American, so that Europe to me is a strange and 
new place that I have come face to face with, with very little chance of 
following any ties back. 

In my anthropological work under Franz Boas I gained integration 
with European scientific thought, because Boas always gave us the 
first reference in German, the second in French, and sometimes one 
in English. My first anthropological work was on the extent to which 
the phenomena of adolescence could be regarded as biological and 
the extent to which they were culturally influenced. That was 1925. 
Since that time I have worked in eight different primitive societies, 


taking to them the problems that were developing in the field of 
human behaviour over that period. During my second field trip in 
1928 I took Professor Piaget's early work to the field and attempted 
to test it out. I am now going back this June, twenty-five years later, 
to re-study the same community, and I will be able to take this next 
twenty-five years of Professor Piaget's work with me. In the same 
way I have used from time to time one set of psychological formula- 
tions and at another stage some other or later development. 

During the last twenty years I have been under the protective aegis 
of the Macy Foundation and all its cross-cultural and cross-discipli- 
nary and cross-everything enterprises. I have been exposed to a very 
large number of conferences such as this, which have brought new 
problems, which one could use to take back to the field. As Dr. Tanner 
said of himself, I may be more useful in answering questions than in 
presenting material, because one of the characteristics of ethnological 
work is that we deal with whole cultures and so many varieties of 
material that it would be rather difficult to anticipate just which kind 
of material would be most useful. At the same time the anthropologist 
can take any hypothesis back into the field and subject it to new tests. 
My major function in these conferences was first to say whether the 
hypothesis was culturally limited or not. Is it physiological to nod 
your head to say 'yes', and did the negative head shake come from the 
child avoiding the mother's breast ? Is it culturally stylized ? Secondly, 
to present hypotheses which come out of the material. In most cases 
primitive material cannot supply proof. We deal with too small 
populations over too short a period of time. We are a hypothesis- 
criticizing, -correcting, and -producing agency rather than a proof- 
producing agency. 


Well, sir, I may add to your confession of cosmopolitanism that I 
have an American mother and an English father, and that I share a 
birthplace with Norbert Wiener, T. S. Eliot, and Harry Truman. I 
was born in Missouri. For that reason my life has been one long 
illustration of the need to 'show me'. 

I am an experimental scientific worker. I started my training in 
the University of Cambridge as a fairly pure neurophysiologist in 
the school of Adrian and Matthews, and I spent five years there, 
studying the detailed neurophysiology of the peripheral nervous 
system. I then had the honour of being delegated by my professor. 
Sir Joseph Barcroft, to work with a Rockefeller Fellow — one of the 
first and few who came from Leningrad — on conditioned reflexes. I 
was given the task of acting as his assistant and becoming familiar 


with the classical techniques of the Pavlovian School. I spent two 
years at that work, having a good background already of neuro- 
physiology. I was enabled first of all to introduce a number of 
modernizations into the Pavlovian technique, to assure myself of the 
essential accuracy of the Pavlovian hypotheses, and to become much 
impressed with the manner in which the Pavlovian workers at that 
time were able to distinguish factors related to personality in their 
experimental creatures, both animal and man. Since that time, as 
you know, that particular aspect of Pavlovian work has been rejected 
and denied by the Soviet authorities, and very few people, I think, 
understand how important the typology of Pavlov was, in the early 
days, to the development and scope of the Pavlovian theories. 

After we had realized that to extend the work in Cambridge would 
cost far more money than was available, I had the good fortune to 
be appointed as a Rockefeller Fellow at the Maudsley Hospital in 
London, where my approach to the human problem was directed and 
inspired by Professor Golla, who was then setting up a new laboratory 
for the multi-disciphnary study of the human organism; I had the role 
there of physiologist. There I was introduced to the study of the 
electrical activity of the brain, which as a physiologist I had previously 
considered to be inaccurate and unlikely to lead to any information, 
the brain being, of course, at that time a most objectionable subject 
of study. I had the opportunity to visit many European centres of 
brain physiology preparatory to setting up our own laboratory. I met 
Berger and Foerster and various other workers in the field of brain 
physiology. Our laboratory was set up mainly for the application of 
electroencephalography to psychiatric problems, but we were very 
soon more heavily involved with neurology, and I devoted a number 
of years to the study of organic lesions of the nervous system. It was 
rather a tough apprenticeship for a physiologist, having to relearn 
neuroanatomy and apply it to what was then an extremely inaccurate 
and troublesome method of study. 

At the end of my period at the Maudsley, just before the War, I 
moved with Golla to Bristol, where I am now, and once again had to 
redirect my ideas towards the more generalized physiology of the 
human nervous system. Our plans were interrupted by the War. 
During the War we devoted our attention mainly to the problem of 
head injuries and epilepsy in Service personnel, but at the same time 
occurred the opportunity to deal with more normal physiology in 
matters quite relevant to the meeting here, that is the problem of 
children evacuated from the cities. Hundreds of ill-behaved and, in 
fact, horrible creatures descended upon us from the slums of big 
cities, presenting one of the most serious problems which my 
country has had to face : the disposal of these young creatures in 


schools, billets, and so forth. We found that the application of 
physiological techniques to the separation, selection, and classifica- 
tion of these children was astonishingly valuable. From that time 
dates my interest in the relevance of the physiology of the nervous 
system to the study of how children grow up, how the influences of 
environment and heredity, nurture and nature, combine to make the 
child as it is. 

These interests have been paramount in my scientific thinking, 
combined obviously with the early influence of the Pavlovian School, 
and I have attempted particularly to quantify methods of study, to 
develop men and machines able to make objective and concrete 
appreciation of the problems which we encounter in this sort of work. 
Tins approach seems to me to have been neglected in the past, and 
ignorance here is hable to produce considerable misunderstanding if 
projected further. 


As my name suggests, I am descended from Italian stock on my 
father's side. I was born in Paris in 1910 from a Burgundian mother 
and a Parisian father. My early teachers were Pieron and particularly 
Wallon. In 1933, having finished my studies at the Sorbonne, I 
decided to go abroad. Chance, or rather political events, led me 
overseas. I intended to go to Vienna to work with Freud, but the 
circumstances were not very favourable so I decided, while waiting 
for the situation to improve, to go to the U.S.A., where I worked for 
six months with GeseU in his Institute at Yale University. I came back 
to France in the Spring of 1934 and it was then that I started work 
directly under Wallon. 

In 1940 1 published my first book, which was devoted to American 
psychology. This was also at a difficult period, and the following 
incident is worth quoting. French censorship, under German control, 
required me to delete all the Jewish names from my book. Naturally 
this was rather awkward. I solved the problem by taking the names 
out of the preface and putting them back in the following chapters ; 
and so I was able to ascertain that the censorship did not read beyond 
the first chapter. At this time, in 1940, 1 was working at the Psycho- 
pathological Laboratory at the Henri Rousselle Hospital. 

Four years ago I took over from Professor Wallon as Director of 
the Laboratory of the Psychobiology of the Child. My present re- 
search bears more particularly on the psychology of epilepsy. Our 
team of workers is dealing with the classical problem of mental 
deficiency, certain types of which we are trying to redefine. It is an 
old problem but always seems new. 


For the last few years my 'hobby-horse' has been the study of the 
foundations of personaHty, using various methods, especially with 

Finally, on the question of method I would like to mention that I 
have attempted to reconcile the clinical method taught by Professor 
Wallon and the statistical, quantitative, approach. There are diffi- 
culties, which are, moreover, exaggerated by doctrinal oppositions 
and opposition between groups which regard themselves as strictly 
clinical or strictly psychostatistical ; but, as I say, I endeavour to 
reconcile these two points of view, both of which seem to me essential 
for the understanding of human nature. 


I am a paediatrician. After my medical examination I studied 
paediatrics in Stockholm, and very soon I was directed by my chief, 
Professor Lichtenstein, towards the field of children's convulsive 
disorders. This was at a time when electroencephalography was not 
at all known as a clinical method in Sweden. It was not possible to 
learn about it at home, and that led to my visiting the United States, 
where I studied at Harvard Medical School, mainly in the chnic of 
Lennox, and also at Johns Hopkins Hospital and the Neurological 
Institute with Hoefer. I went back to Sweden again and continued in 
this field. However, I found it very difficult to judge pathological 
conditions, knowing almost nothing at all about the normal in 
children. It has therefore been a constant struggle during the years 
to find out what is normal in children, especially in the field of the 
electrical development of the brain. My interest has been mainly in 
that direction and I have tried to work in this field from the electro- 
encephalographic point of view. I am, however, still a pure paediatri- 
cian, and I have still my main interest focused on work with children 
suffering from convulsive disorders. 


I began my studies in Picardy, in the north of France, and continued 
at the Faculty of Medicine of the University of Paris. There I was 
particularly interested in the study of the nervous system of the adult 
and the child. One of my best-liked tutors. Professor Baudouin, was 
the first in France (in 1936) to become interested in electroencephalo- 
graphy. In 1939 I was already thinking of concentrating on neuro- 
physiological research. 

In 1941 I started working regularly in Professor Baudouin's 
laboratory, where I studied mainly the electrical activity of the brain 
of sick children, with the assistance of Professor Heuyer and Professor 


Debre. I was lucky enough in 1945 to work for a year in the U.S.A., in 
Professor Detlew Bronk's laboratory at Philadelphia; there I was able 
to become familiar with the techniques of neurophysiology. Although 
these techniques had perhaps not entirely escaped notice in France 
during the war, little was known of the details. I was able to study 
particularly the polarographic recording of oxygen and apply it to 
the study of cortical metabolism. 

Since 1946 I have been with the Centre Nationale de la Recherche 
Scientifique and have concentrated on the electrophysiology of the 

I went back to the U.S.A. in 1952 to work with Professor McCulloch 
in Chicago, and since my return I have again held the post of Head 
of the Laboratory of Electroencephalography and Applied Neuro- 
physiology. As a daily routine in this laboratory we make electro- 
encephalograms of about twenty persons presenting different lesions 
of the brain. Having made most interesting contacts at the Salpetriere 
in a neurological milieu which is already long established, I want to 
profit by the atmosphere which exists there to base on it research in 
human neurophysiology. 

Regarding the research which I have followed or carried out 
myself, I should like to mention that in our laboratory we are 
attempting much more than in the past to 'define' the normal 
individual. Since we are constantly working with sick persons we 
cannot always know where the pathological begins and the normal 
ends. What is the normal individual? What is the normal child? One 
might ask whether normality exists. 

Among the groups of people we try to study I should Uke to refer 
to a very restricted one, of pilot apprentices. In order to show the 
difficulties we are faced with I would mention that individuals of 
twenty to twenty-four years in this group who appeared absolutely 
normal according to screening, definitions and tests, seem to us, used 
to observing adults, to be immature as regards the electrical activity 
of the brain. I mention this fact before we start on our work in order 
to ask each of you to underline as far as possible what you consider 
normal and to tell us what importance the definition of normal has 
in the fields you represent. How can we obtain this definition ? Why 
should we obtain it ? 


I think I might, with due apologies to Dr. Lorenz, describe myself 
as some sort of bird of migration. I was born and studied medicine 
in Germany and took my degree at Leipzig University, and then I 
started on my first Transatlantic migration, to the Argentine, from 


which I came back in order to concentrate on neurology and 
psychiatry at the University of Munich, at the University of Paris, 
and for some time also in Zurich. Eventually I wound up being a 
lecturer in psychiatry and neurology at the University of Cologne, 
where I remained until Hitler came into power and, as Hitler and I 
couldn't see eye to eye on many things, I decided that I'd rather 
withdraw to South America again, and so I settled in 1933 in 
Argentina. Since then I've commuted between South America and 
Europe. Eventually I became an Associate Professor of Psychiatry in 
Buenos Aires, and somewhat belatedly the University of Cologne 
also conferred the title of Professor on me. 

As to professional experience, there I think I am a bird of migra- 
tion too. I started on neurological hnes, did a certain amount of what 
is called solid research on the neurological aspects of psychiatry, 
until I found out that I couldn't do without some other sort of train- 
ing; so I underwent my personal psychoanalysis, and since then have 
been commuting between neurological and psychoanalytic psychiatry. 
I feel that birds of migration in a way have some sort of stability, 
they always fly by the same route, and if there are two schools of 
psychiatry — one in which one tries to explain everything in physio- 
logical terms, and the other one where everything is couched in 
psychological terms — I have always been most interested in the gap 
in between, which is in my opinion neither quite as wide as some 
people seem to think, nor quite as narrow as some others beheve. 
I think it is a most fascinating task to see how things can happen in 
physiology through psychological stimuli, and how psychological 
events use physiological channels in order to manifest themselves, and 
in this context lately I have been interested in the 'gap' in epilepsy 
and the convulsive states generally, and in the so-called psychopathic 
personalities, two subjects which are particularly closely related to 
some of the most outstanding neurophysiological problems in 
children. Lately also I have been particularly interested in the impli- 
cations of speech pathology from the point of view both of psychology 
and brain pathology. 


I am a Londoner born and bred. My father's family came from 
Yorkshire, my mother's from Wales, and my father was a surgeon. 
I had rather a wayward youth, inasmuch as I toyed with being a 
sailor and went to the Royal Naval College ; I then took up medicine, 
then switched to child psychology, and it was unfortunate that it 
was at my most wayward that my father died. This was at a point 
when, after reading natural sciences, medicine, and psychology at 


Cambridge, I decided to give up medicine in order to take up 
education. I spent twelve months in one of the progressive and free 
schools, which was a very valuable experience, because I saw a 
number of disturbed children at first hand, I lived with them, indeed 
I had to look after them, and I met there the first 'affectionless 
character' of my career. 

Fortunately, I was very wisely advised at this point to finish my 
medical training and train in psychoanalysis. So I went to London, 
to University College Hospital. Then I speciahzed in psychiatry at 
the Maudsley Hospital and continued my training in psychoanalysis ; 
I finished that and took up child psychiatry and child guidance. 
Between 1936 and 1940 I was concerned with child guidance and it 
was really at that time that I became convinced in my own heart that 
certain events of early childhood were of critical importance in 
determining personality development — particularly the child's 
relationship to his mother, and the mother's unconscious attitude to 
the child, based on her own childhood experiences. I ought to say 
that my concentration on the mother-child relationship was largely 
due to the influence of Melanie Klein. Now, I was eager to make 
scientific these chnical observations on mother-child relationships, 
and I seized on the particular relationship between the experience of 
a child being separated from his mother and the psychopathic 
affectionless character, not because it was the most important, 
but because it seemed to me the most concrete and the simplest to 

Then came the war. I spent five years in the army as an army 
psychiatrist, and much of my time was spent in officer-selection 
work. I received a post-graduate education in psychology in the 
army and a training in research method. I also learnt that the way 
to get people of diverse backgrounds and disciplines and outlooks 
to work together was to give them one single task. 

After the war I was offered a full-time post at the Tavistock Chnic, 
where I have been for the last seven years, in charge of the Child 
Guidance Department ; there I have had one foot in the clinical field 
and one foot in research, trying to 'scientificate' the clinical field. I 
think when I returned to child guidance after the war I did so with 
some doubt as to whether my clinical convictions of pre-war days 
were going to stand up to further scrutiny, and I was rather delighted 
to find that they all did, or at least they seemed to. 

I returned very swiftly to my hobby-horse, mother-child separation. 
It seemed to me that it was one of the few islands of dry ground in a 
rather swampy scientific field, and that one had here a definable 
experience which demonstrably could sometimes produce a particular 
type of personality outcome. I have stuck very rigidly to it, with two 


or three purposes in mind. The first has been to substantiate a claim 
that all child psychiatrists make, that these early experiences between 
the parent and child are really as important as we think they are, in 
contrast to the view of many who ridicule it. The second purpose is to 
make clear that here is an aetiological factor calling for preventive 
action in the mental health field, something clear cut and concrete 
that people can get hold of. The third purpose is research ; I felt that 
we had here a scientific phenomenon to the study of which many 
techniques could be brought which might lead to a unification of 
different points of view. My own techniques were those of psycho- 
analysis and child psychiatry but I hoped to bring to bear on this 
one single problem a variety of disciplines. It was my good fortune in 
1950 to be invited by W.H.O. to read the literature on this subject 
and really get to grips with it. Since then I have been interested in 
getting hold of any scientific knowledge which seemed to lead to an 
understanding of why it is that that particular sort of experience, 
the child being separated from his mother for many months, could 
not only have an effect on the character but have a permanent effect. 
What is the cause of the permanence ? Well, that led me, amongst 
other things, to be interested in Professor Lorenz's work; the 
phenomenon of imprinting at once struck me as possibly important 
to my work. Whether it really has anything to do with the effects of 
separation we shall see. The other thing that fascinated me in his 
work was the mother-child relationship of animals. The mother-child 
relationship is manifestly an example of instinct, in the ethological 
meaning of the word, and it is also at the centre of psychoanalysis. 
Other things which have been interesting me more recently have 
been the phenomena of behaviour under stress and experimental 
neurosis because, here again, one is faced vdth peculiar forms of 
learning which have a remarkable quality of persistence. My interest 
in this Study Group is the hope that some other disciplines could help 
me in my quest. I must confess to a rather one-track, one-problem 


I am born and bred Austrian, from near Vienna. My scientific 
career, rather hke that of Dr. Mead, started at the age of five, when 
I got a nest of ducklings as an Easter present, and I may say that the 
ducklings and I became imprinted upon each other, and curiously 
enough the history of my teacher Heinroth, who has done most of his 
work on Anatidae, begins in exactly the same way, with a goshng, 
not a duckling, but also at five years. Well, that may be a coincidence. 
Then, at about twelve years, I became acquainted with Charles 


Darwin, by means of a popular booklet (Darwin, 1872), and from 
then on I became a passionate evolutionist, and besides started to 
become a comparative zoologist, which my father, who was an 
orthopaedic surgeon, tried to prevent by ordering me to study 
medicine. This afterwards saved my life, when I was in Russia. 
Nevertheless, I resented it strongly at the time, and when I was a 
graduate student, I attached myself to Professor Hochstetter, who 
was the best morphologist and comparative anatomist living, and 
decided to become a comparative anatomist, not changing my 
preference for evolution, though. I did not give up keeping live 
animals, but regarded it as a hobby, a plaything. Now, at Hoch- 
stetter's chair, I learnt how one ought to proceed in investigating 
and reconstructing the course of evolution. Then I learnt among 
other things one fundamental sentence, which my teacher Hoch- 
stetter used to repeat again and again, 'Primitive animals do not 
exist, only primitive characters do.' 

Then I made my discovery; it wasn't mine, it was actually Charles 
Whitman's, but I didn't know it. I discovered instinctive movement, I 
discovered for myself that innate movements are just as conservative 
characters in the species or genus as are any claws or bones. The 
museum zoologist is apt to think that what is most resistant and 
keepable in the museum must be so in evolution, but it isn't true. 

Then, when I was just full of the first discovery, I discovered 
Heinroth, and I saw that he knew the same thing, and I found 
out that comparative morphology of movement might be worth 
while studying. At the same time I was not, and Heinroth was not, 
interested in the physiology or in the psychological importance of 
instinctive movements. We were interested in comparative characters. 
We were interested in finding more and more characters that could 
be used in reconstructing evolution. And simply because there were 
not enough morphological characters available, we proceeded to 
spread our search to behaviour patterns and found that these very 
often are even more conservative and reliable landmarks of phylogeny 
than the colours of plumage or the form of bones. In view of the fact 
that instinctive movements were discovered in this particular way it 
always seems somewhat paradoxical to me that there still are people 
who try to deny their very existence. 

And now on the strength of Heinroth's authority, I decided to 
make my profession out of what had been my hobby up to that time 
in the study of animals, and I quit my position as an assistant in the 
anatomical institute, and became docent in Vienna for comparative 
psychology. At the time that was impossible, because in Vienna the 
very Catholic regime prohibited animals having souls and therefore 
a psychology of animals was impossible ; and so I got a lectureship 


for comparative anatomy and — hush-hush — psychology. I have never 
deHvered a lecture on comparative anatomy in my life. Then I 
became professor of psychology in Koenigsberg, together with the 
pragmatist philosopher Eduard Baumgarten, a disciple and ardent 
admirer of John Dewey, Together we had a great number of good 
fights with Neo-Kantian philosophers, and, in general, it was a 
very nice collaboration between philosophical anthropology and 

For a very short time I became a soldier and then was recruited as 
an army surgeon because I had a full doctor's degree of medicine. 
From then on I worked as a psychiatrist and neurologist, having been 
promoted to the rank of an Unterarzt, which is a sergeant. I was for 
two years in Posen, under a very intelligent teacher, Dr. Weigl, who 
was interested in the study of neurosis. It was interesting to me that 
neurosis developed in people who had an instinctive inhibition towards 
kiUing. Such people were my patients because they couldn't kill any 
more, and what struck me as most gruesome was that there were so 
few of them — one would have expected many more. Then I was 
transferred to the front and immediately caught by the Russians, and 
then I went on being a neurologist and a psychologist in Russia, 
without any interruption. There I was treated very decently indeed 
for four years and wrote a volume of my textbook (Lorenz, in press) 
there, and was able to bring it home. Then I came to Vienna and 
again developed my little Institute of Comparative Ethology in 
Altenberg, with the help of the Austrian Academy of Sciences, which 
was financed by the English poet and writer, J. B. Priestley, who gave 
all the royalties of his plays and publications in Austria to the 
Austrian Academy of Sciences. My work went on very prettily on 
that but the family finances did not; we had to sell one piece of 
ground after the other ; Altenberg became smaller and smaller. Then 
I got a call to the University of Bristol, where I was to do compara- 
tive ethological work on the beautiful collection of ducks and geese 
collected by my friend Peter Scott, and simultaneously I got an 
appointment by the Max-Planck-Gesellschaft to take over a newly- 
estabhshed Institute, and I decided to take the latter, although after 
very much hesitation. 

Our present work tends to become interested, as you see, in 
knowing more and more about less and less ; the problems begin to 
specialize more and more and therefore I think it is very good for me 
to take part in this meeting to widen my interests again. It is a great 
satisfaction to me that comparative ethology is something that child 
psychologists and psychiatrists begin to get interested in, because 
what we do is only to create a base ; whether you can use it is some- 
thing beyond my knowledge. Well, I hope you can. 



May I ask one question? I would like to know when the term 
'ethology' was coined, and by whom. 


The term ethology was created by Heinroth, who, with Whitman, 
was the pioneer of that science. It's an interesting fact, though, that 
neither knew of the other's existence, still less about his work; 
Heinroth called one of his first and most important papers 'Beitrage 
zur Biologic, insbesondere Psychologic und Ethologie der Anatiden' 
(1911). The subject of this paper is constituted by the innate activities 
and reactions of the birds in question. Tinbergen took over the term 
ethology, which I consider rather a pity, because it creates mis- 
understandings with psychologists and philosophers. Neither Hein- 
roth nor Tinbergen cared in the least about human psychology and 
philosophy, sciences in which the words ethos and ethics have a very 
different meaning. So they did not mind the ambiguity of the word 
ethology. But I think it is too late to do anything about it now, we 
are called, we are branded, 'ethologists'. I never use the term ethology 
in German, though; I say "Vergleichende Verhaltenslehre' . The 
German ending -lehre has the advantage over the English -ology that 
you can join it on to practically anything. You could not say 'Com- 
parative Behaviourology' in English — and that is what probably 
influenced Tinbergen. But the word ethology is really awful, I concede 
that to any psychologist. 


I am Swiss, and spent my childhood and adolescence in eastern 
Switzerland. Being the daughter of a zoologist I amused myself, as 
naively as many other children do, rearing various animals, including 
tortoises, without in the least suspecting the role Dr. Grey Walter 
would one day make them play in cybernetics. I went to the Univer- 
sity in Geneva where I studied under Professors Claparede and Piaget. 

At the Institute of the Sciences of Education in Geneva it is common 
practice to throw students into the water to teach them to swim. So 
from the very beginning M. Piaget asked me to participate in his 
work and assigned me a piece of experimental research; it was a 
matter of studying how a child forms the idea of physical conserva- 
tion. M. Piaget proposed that I should do this by dissolving pieces of 
sugar in water to see whether the children believed in the destruction 
of matter or supposed that it was conserved, and to see what atomistic 
intuition they had (Piaget and Inhelder, 1941). 


At the beginning of the war I was asked to estabhsh in German- 
speaking Switzerland a centre for the psychology of school children 
with the main object of discovering and diagnosing cases of mental 
retardation. The psychological research on development I had carried 
out previously with a purely scientific and theoretical aim was then 
most useful to me. The stages in the genesis of the concepts of 
conservation supplied me with a scale of development on which the 
phenomena of mental deficiency showed as delays and fixations 
(Inhelder, 1943). In my case-finding investigations, which led me 
from one school to another, I resolutely devoted to a scientific cause 
the few pieces of sugar allowed per month at that time ! 

In 1943 I was recalled to the Institute, to act firstly as 'Chef des 
travaux', then as 'Chargee de Cours', and from 1948 as Professor of 
Child Psychology. My main activity, however, is still directed towards 

With a group of assistants I train the young students in the use of 
scientific and cHnical methods by allowing them to participate in our 
work. In recent years we have concentrated on the genesis of spatial 
concepts (Piaget and Inhelder, 1946; Piaget, Inhelder and 
SzEMiNSKA, 1948) and of physical and mathematical concepts 
(Piaget and Inhelder, 1951). At the moment we are hoping to 
conclude a series of studies on experimental and inductive reasoning 
in children and adolescents (Inhelder, 1948, 1951). 

In recent years we have been working in a highly specialized, 
restricted area: that of the genesis of intellectual functions in children. 
This specialization was a result of the need to push our research 
deeper, but until it rests on a neurological basis and as long as we 
remain ignorant of the emotional and social background of the child 
I consider it to be hanging in mid-air. As I wish to incorporate our 
results in a comprehensive study I welcome the possibiUty of colla- 
borating with Dr. Monnier and Dr. Odier. I am particularly glad of 
the opportunity of participating in this meeting and hope not only to 
obtain information but also to make new plans for research. 


I was born at Neuchatel, in Switzerland. I was less precocious 
than Dr. Mead and Dr. Lorenz because I was already fifteen when 
my first work was published. This work dealt with a special field of 
zoology — the study of terrestrial molluscs. 

The way I came to study child psychology was far from orthodox. 
I had studied natural sciences, and my doctor's thesis dealt, of course, 
with molluscs. What interested me most were the problems of adapta- 
tion, of the relation between an organ and its environment and the 


problem of variation as a function of environment and as a function 
of structure. But while I was preparing my doctorate in zoology I was 
taking a very lively interest in the problems of knowledge, of 
epistemology, of logic, and the history of sciences, etc. At the same 
time I was highly suspicious of philosophers who I thought had 
treated the problem of knowledge in a fashion that was far too 
speculative and not sufficiently experimental. I then considered 
devising a genetic theory of knowledge, studying knowledge as a 
function of its growth and development, so I felt I should read 
psychology. I thought I would spend four or five years studying the 
development of logic and the intellectual functions in the child and 
the growth of intelligence during the child's development; these 
studies have lasted more than thirty years and are not yet finished. 

To begin with I concentrated mainly on the problem of logic in the 
child, then on the concepts that are basic to science : number, space, 
time, etc. My method of studying logic in the child was much too 
verbal at first, dealing particularly with the relation between thought 
and language. Gradually I discovered that for this study it was essen- 
tial to go back to the actions themselves, to the reasoning which is 
carried out not through language but through manipulation of 
objects. Starting with my books on the first and second years of child 
development my technique has always been to study reasoning through 
objects set up so that the child could make certain experiments. 

This study of the child's actions brought me to a conception of 
logic based on operations — an operation being considered as inter- 
nalized action which becomes reversible, that is to say can be 
carried out in both directions, and finks up with others. In the 
sphere of intelligence operations always constitute whole structures, 
rather like the Gestalt in the sphere of perception; the structures 
being, however, larger, more mobile and essentially reversible, and 
capable of co-ordination. For several years we have studied these 
structures in the infant, then in the child of seven to twelve years, and, 
finally, thanks to the recent work of Mile Inhelder, in adolescence. 
These structures are of great interest. As one arrives at a certain 
degree of generafization in the study of operative structures one finds 
again the fundamental structures of mathematics; algebraic struc- 
tures such as the group, or structures based on the idea of order, 
such as the lattice concept, and topological structures, etc. We are 
studying the achievement of these structures in the adolescent and 
their development during childhood, and they seem to give some hope 
of co-ordination between psychology and neurology. It is evident 
that such general structures are based on the activity of the brain. 
Although perhaps no neurological contact is possible at present, 
except for attempts — such as those of Pitts and McCulloch — to 


apply logical structures to neuron structures, I think nevertheless that 
now we can go further in developing the comparison between the 
various cybernetic models and the activity of intelligence, as we have 
tried to define it. Moreover the attempt to discover all possible 
connexions between these fields comes within the frame of reference 
of this meeting. 


Dr. Mead told us that she began her career at the age of three. By 
comparison I feel like a newcomer, because my entry into the corpora- 
tion dates only from my Oedipus complex, at the age of five or six. 

I had the rare good fortune to have a father who was most 
interested in all the questions I asked and took the trouble to reply. 
I went through the phase called 'questionnisme' in French. As you 
can imagine I gradually realized that my good father did not know 
everything, that many problems remained unsolved, and I thought it 
was the duty of my generation to study them more closely. That is 
the origin of my vocation, for all my questions were connected with 
the thoughts, intentions, and opinions of grown-ups. 

I will pass over a few years and come to an important event which 
occurred during my medical studies. Professor Flournoy gave a 
summer course in psychology and I had the privilege of making his 
personal acquaintance. I would mention that Professor Flournoy was 
the first person from a French-speaking country to become interested 
in a certain method, a certain psychological conception, taught at 
that time by a Dr. Freud in Vienna. 

Later, after the first war, I spent two years in the department of 
neurology at the Salpetriere, mainly under Professor Pierre Marie and 
Professor Souques, who was Charcot's first pupil. I worked in a 
section for wounded soldiers where cases of mental shock were 
unfortunately fairly frequent. I saw several cases of post-shock 
syndrome psychoanalysed for months or even years with truly remark- 
able results. The patients recovered despite the opinion of the 
psychoanalysts, who considered the aetiology of the syndromes to be 
purely psychogenic. Because of this period spent in Paris I always 
think of the possibility of faulty diagnosis and consider the delicate 
question of physiogenesis or psychogenesis. 

Later I studied in Vienna. It was then that I deviated. I was going 
to be led astray into psychiatry. One day, looking through the pro- 
gramme of courses, I found that Freud was giving a course on the 
theory of instinct. Remembering my conversations with Flournoy 
and Claparede, I decided to follow this course. There were ten or 
eleven students — twelve on a good day. 


After that I worked in the Psychoanalytic Institute in Berlin. I was 
psychoanalysed and received psychoanalytic training. My teachers 
were Freud's first pupils : Sachs, Abraham, Rado, and even Jung. 

I spent another period in Paris where I had been asked to contri- 
bute to the creation and organization of the Institut de Psychanalyse 
and the Revue Fran?aise de Psychanalyse. 

Gradually, however, I became aware of what I might call a lacuna 
in Freud's theory — the psychology and activity of the ego. At that 
time it was very bad form to mention the ego in a psychoanalytic 
discussion, but I discovered that this evaluation was only a resistance, 
or rather a defence mechanism; actually my colleagues certainly felt 
that something was wrong. It must be admitted that data on the ego 
in Freud's theory were precarious, insufficient and sometimes even 
contradictory. Therefore in Paris, in order to try to bridge this gap, I 
turned to the work of the genetic psychology school and plunged into 
the works of Piaget. I was immediately struck by the many connexions 
between neurotic thinking and all the mechanisms described by 
Freud, and infant thinking as described by Piaget and his followers. 
In my course at the Institute of Psychoanalysis I tried to establish 
links between stages of development of thinking and intelligence as 
described by Piaget and instinctive-affective stages described by Freud. 
I immediately ran into great difficulties, I found that it was exceedingly 
difficult to connect these different stages, as if the child adopted 
either the theories of Freud or those of Piaget, developing in one 
direction or the other. 

However, there are two points on which I think a correspondence 
exists, and which play an exceedingly important part in the psycho- 
biological or biologico-psychological development of the child, parti- 
cularly in this highly complex evolution which should culminate in 
the sociahzation of the individual. These two points are: firstly, every- 
thing connected with the mechanisms of the super-ego and, secondly, 
everything connected with the well-known Oedipus complex. The 
latter, in my opinion, is a most important and critical phase in the 
development of the individual ; I would almost say a second critical 
phase, admitting the great importance of the first critical phase so 
well described by Dr. Bowlby. The Oedipus complex is not a sign 
of illness ; if it develops normally it is, on the contrary, a factor making 
for balance and development. 


It is perhaps relevant to my career that I was born in South Africa 
and lived for several years of early childhood in South Africa, and 
had an African nanny. Then I went to England, was educated there, 


and so to hospital and took a medical degree about 1926. After 
qualification, I went to Kenya Colony as a medical officer of the 
Government and for nine years worked as a general medical officer 
throughout all parts of the Colony. I did medicine and surgery and 
midwifery and saw the people in their homes and in the hospitals 
and the out-patient departments of the little hospitals throughout 
the Colony, and throughout those nine years I must have seen a 
number which runs into six figures of the population of Kenya. 

Round about 1937 the Colony had developed to the point where 
it was considered necessary for a medical man to be permanently 
attached to the Mental Hospital. There had been for some time before 
that a mental hospital, but it had been regarded only as a place where 
any doctor who happened to be near at hand could look in from time 
to time and deal with emergencies. There were two schools of 
thought in this matter — one school thought that they should get an 
expert from England who could study the African and the African 
language after he arrived; the other school of thought felt that they 
should get somebody from the Colony who knew the people and the 
language and could take his degree in psychiatry at a later time. The 
first school won the day and an expert was brought from England, 
but this arrangement did not last very long, and nine months later I 
was called upon to fill the breach. I don't know to this day why I was 
chosen, but I accepted the post and after taking it on I became more 
and more intrigued and interested in the work ; but I had no oppor- 
tunity to return to England for many years, because the war inter- 
vened, and I had to do my psychiatric study of the people on the basis 
of textbooks, and of talking to occasional people who knew some- 
thing about the subject and who were passing through the country, 
and I found myself in very great difficulties over diagnosis, and 
finally decided that I must be a very bad diagnostician. 

However, the opportunity came in 1946 to return to England and 
take the Diploma of Psychological Medicine in London, and then it 
dawned upon me that my difficulties were partly due to the fact that 
the pictures of disease as described in European textbooks of 
psychiatry are not at all completely applicable to the pictures that 
one sees in Africa, and this has been the theme of my interest for the 
last several years — are these differences real and if so what accounts 
for them. In regard to this meeting on the psychobiology of the child 
I should say that it seems to me that in many ways the African, as 
an individual, stops short at the second stage that Professor Piaget 
has described and I would like to know why this is so. 



Physical and Physiological Aspects 
of Child Development 


I am sorry to say that during the next fifteen minutes I am going to 
make a considerable number of statements without giving you the 
detailed evidence for many of them. It seems to be the only way I can 
effectively get across what I have to say, which will really constitute 
my feelings about the growth process, based entirely on morpho- 
logical and physiological growth and not on the psychological aspects, 
about which I have no personal experience. These ideas may or may 
not clash with the ideas coming from the psychological side. 


I think anybody who has studied the growth curves of infants and 
children must be struck by the fact that the whole affair is quite extra- 
ordinarily regular. Different external dimensions and different organs 
grow at different rates, because the head end of the foetus develops in 
general earlier than the tail end; thus the head after birth grows 
slowly, the legs, less advanced, grow quickly. But as far as is known 
each dimension follows a perfectly regular rate-of-growth curve with 
no breaks or spurts as long as the environment is optimal. The one 
possible exception to this statement is the mid-growth spurt, an ac- 
celeration which may occur and be confined to breadth and width 
dimensions between about five and a half and seven and a half years. 
The evidence for its existence is dubious, and the subject needs re- 
studying on more exact longitudinal growth data. At adolescence a 
striking growth spurt occurs in all external dimensions except fat. 
Immediately prior to this adolescent spurt there is a wave of fat 
increase, followed during the spurt by fat loss. 

All this is even true of the growth of single individuals, which after 
all is almost more than one could demand. There are bound to be 
variations in any process, yet they are not sufficiently large, short of 


FIG. 1 



(From Scammon, 1930, The Measurement of Man, Univ. Minnesota Press) 



Thymus, Lymph-nodes, Intestinal lym- 
phoid masses. 


Brain and its parts. Dura, Spinal cord. 
Optic apparatus, many head dimen- 


Body as a whole. External dimensions 
(with exception of head and neck). 
Respiratory and digestive organs. 
Kidneys, Aorta and pulmonary trunks. 
Spleen, Musculature as a whole. 
Skeleton as a whole. Blood volume. 


Testis, Ovary, Epididymis, Uterine 
tube. Prostate, Prostatic urethra. 
Seminal vesicles. 

6 6 JO 12 14 16 16 20 
Age in Years 

Curves drawn to same scale by plotting as percentage of adult (20-year-old) values 

at successive ages 

2 4 

malnutrition and suchlike, to disturb the regular genesis of the 
curves that one sees. Curves of a smaller period may be imposed on 
the general curves. For example, there are differences between the 
seasons; a child grows more in height in the spring and more in 
weight in the autumn. This is very well established and the magnitude 
of the effect is considerable, but again it does not disturb the general 
underlying regularity of the process. I cannot help feeling there is a 
certain inevitabihty about growth which forcibly reminds one of the 
sort of development that, for example. Dr. Lorenz is interested in. 
The mechanism unwinds; it is not as though it is being pushed 
particularly to do so, but it just unwinds unless you get in the way 
and stop it. 
Fig. 1 is a very famous illustration from Scammon (1930) that many 


of you will know. It shows some growth curves from birth to age 
twenty. These four curves are of different tissues, to demonstrate that, 
though each different part of the human organism has a regular 
curve, these curves are not all the same. You will see that the growth 
of the brain and the spinal cord reach the adult level early. What 
Scammon calls the general type of curve characterizes the growth of 
most of the external dimensions such as shoulder breadth, width of 
the chest, height and, approximately, weight. These dimensions grow 
fairly fast after birth, then they slow down for a while and then have a 
great spurt at adolescence. There are two other forms of growth 
illustrated in the figure ; there is the growth of the genitalia, which lie 
dormant until adolescence and then suddenly catch up with the rest 
of the organism; and there is the growth of the lymphatic tissue which 
is rather considerably different from the others. The lymphatic tissue 
grows to a supra-adult magnitude and then at about adolescence it 
decreases all over the body, even in organs which consist chiefly 
of other tissues. 

The same events are shown in Fig. 2 in terms of velocity. These 
velocity curves, that is curves of rate of growth, seem to me often 
more informative than distance curves when discussing physical and 
physiological growth. One can see from this figure that after birth 
one grows more and more slowly. The brain practically stops growing 
in magnitude by four or five years old, for example. The growth of 
the external dimensions — body weight is plotted here, but it could 
equally well be stature, or some other measurement — decreases also 
in the same way, but then has the adolescent spurt. The weight of the 
testes shows a much greater adolescent spurt. Lastly, the thymus and 
the lymphatic tissues after their initial decrease actually have a 
negative velocity — as we have seen in the previous figure. 

In Fig. 3 are the growth curves of the hip width of a group of girls 
studied more or less longitudinally. Above is the distance curve, 
plotting the average width of the hips each year; below is the 
velocity curve, showing the rate of growth in hip width. The time of 
fastest growth for a dimension like hip width is before birth, at about 
six to seven intra-uterine months; after this the velocity decreases 
steadily except that the decrease is interrupted twice. There may be an 
increase of velocity at about five and a half to seven and a half years, 
the 'mid-growth' spurt (Tanner, 1947). Then there is the adolescent 
spurt about whose existence there is no question whatever. 

There is one other point worth mentioning before I leave these 
figures ; the mechanism of form change. We do not look exactly the 
same as we did when babies, and this form change is due to some parts 
growing faster than others at various times. For example, supposing 
one believes in the mid-growth spurt; there does seem to be a 


FIG. 2 



(From Scammon, 1930, The Measurement of Man, 

Univ. Minnesota Press) 









Tesfes weigfit- 

Thi/mus uxeight 

—--'Brain caieigFil- 




































\ 1 




















'V, ' 





6 10 /^ 
fige in years 

14 16 18 20 


FIG. 3 








Bitrochanteric Diam 

J2 5 

1 2 3 4 5 6 7 8 9 10 11 12 15 H 15 16 17 

Adolescent Spurt 

X Midgrowth Spurt 

\ r 

1 2 5 1 5 6 7 6 9 10 11 12 15 11 15 16^ 17 
AGE > 

After Tanner (1947); Data from Simmons (1944) 

difference between growth in length of the body at that time and 
growth in breadth. Fig. 4 illustrates this. It shows that the velocity 
of trunk length, arm length, and leg length (above) continues to fall 
during the mid-growth period while that of hip width, chest breadth, 
thigh circumference (below), increases. 


FIG. 4 






- 3 

Length \ 

1 I I I I I III I I I 1 1 1 1 1 ^-l-^ 1-^,; . . 

] Z 3 4 5 6 7 6 9 10 11 12 13 \4\ 15^ 16 17 


After Tanner (1947); Data from Simmons (1944) and Boynton (1936) 


Nervous System 

A few words now about the development of the nervous system. 
We know very Httle about this. We lack even the simplest anatomical 
facts about the development of the human brain, such as, for example, 
growth curves of the nuclei of the thalamus or hypothalamus. The 
external growth of the brain approaches completion earlier than any 
other bodily dimension, and there is little or no new nerve cell forma- 
tion after birth, or even for some time before. At least this is true of 
the cerebral cortex, which is the only part about whose growth we 
have any good information. Very little indeed is known of the 
growth of the subcortical structures ; the times of appearance of the 
various nuclei and tracts are for the most part unknown. Presumably 
there is a regular sequence in their development comparable to that 
of the ossification centres. There is the same ignorance about the 
development of the sense organs. 

By nine months after birth, the brain is 50 per cent of its adult 
weight, and by two years it is 75 per cent. At about three intra- 
uterine months the cells become organized into the layers of the 
cortex, and at about six intra-uterine months the layering and the 
appearance that we are familiar with in the adult is present. After 
that the cytoplasm of the cells grows, and the cell processes enlarge, 
but no new cells, or very few new cells, seem to appear. One can only 
suppose, of course, that after this time new tracts are beginning to 
function, connexions are being made, and neuroglia and blood vessels 
are growing. Conel (1952) has studied the cerebral cortex from birth 
to six months and makes one very interesting point: the primary 
motor area is the most advanced of any part of the cerebral cortex, 
he primary sensory areas are the next most advanced, and the 
primary auditory and visual areas follow them. In the case of each 
sensation the association areas are less advanced than the primary 
receptive areas. The cingulate gyrus, the hippocampus, and the 
insula, which are concerned particularly in circuits which include 
subcortical structures, and as a rough generalization can be said to 
be concerned mainly with the emotions, are developmentally behind 
the sensory areas at that time. 


What do you mean by 'advanced' ? 


Conel takes the following grounds. Increase in the width of the 
horizontal layer of the cells ; decrease in number of nerve cells per 
millimetre; increase in size of nerve cells; increase and differentiation 


in the chromophil substance and in the neurofibrils. In principle 
Conel's point is that any particular part of the cortex changes in 
several ways as time goes by; and when he says that one area is more 
advanced than the other, he means that the changes are further along 
in time. 

Is there no reference whatsoever to the myeUnization of the fibres ? 


Yes, he takes nine points altogether as his criteria, and the ninth 
is myelinization, which he regards, if I understand him, as not one 
of the most important. 

That brings us to the end, as far as I am aware, of the information 
on the growth of the nervous system. It does not appear to have an 
adolescent spurt Uke most parts I mentioned. The skull, as a matter 
of fact, does have quite a little adolescent spurt, but it is said that the 
brain does not — I do not see how anybody can tell, with the studies 
that have been done up to the present time. 

Physiological Development 

Now I want to go on to endocrinological and biochemical develop- 
ment. There is remarkably little to say, once again. Indeed of all our 
lacks in the study of the growing child, this is perhaps the greatest. 
There are no longitudinal studies of hormone excretion or hormone 
blood levels, except over the span of a single year. There are a few 
cross-sectional studies from which the course of events is known very 
roughly. So far as this limited information goes, there is no striking 
change in hormonal or other internal environment from six months 
up to puberty. Some parts of the pituitary, some parts of the adrenal 
cortex, and the gonads do not function until puberty; other hormones 
seem to be secreted in proportion to the size of the growing child. 
Striking changes in the internal environment occur at adolescence, 

One of the more interesting things that I can tell you, I think, is 
about the adrenal cortex which currently is believed to secrete at 
least three, and possibly more, sorts of hormones. One of these 
sorts, the gluco-corticoids, or 11-oxysteroids, hormones which are 
concerned in the response to stress, are secreted from very shortly 
after birth at the same intensity as in the adult, per surface area: 
there is less in absolute amount than in the adult, but then the child 
is not so big (Talbot, Wood, Worcester, Christo, Campbell and 
Zygmuntow^icz, 1951). 



Is that based on urinary excretions ? 

It is based on urinary excretions only. 


So we are reaUy still ignorant as to the actual rate of the formation, 
other than by this indirect method ? 


That is perfectly true. Although the blood studies that have been 
done do appear to concur with the twenty-four-hour excretion studies, 
one must be careful how one argues from twenty-four-hour excretions 
back to such things as blood levels and rates of secretion. 

The neutral 17-ketosteroids, which are in general end-products 
of androgenic hormones produced in the adrenal cortex and testis, 
appear to start being excreted about nine or ten, in the male as 
well as in the female. Some people have referred to this, rather 
inelegantly, I think, as the 'adrenarche'. This is an inaccurate term 
because, as I said, one part of the adrenal has been going hard at it 
ever since birth. About other hormones we have no information 
whatever, and I ought to stress the fact that even the information I 
have given you is very insecurely based and subject to change as 
methods and data improve. 

Now I want to consider interrelations during development. The 
functional state of a nervous centre may often depend on the internal 
environment of the body. For example, the nervous pattern of 
copulation and orgasm is complete in man by early childhood, but is 
not normally stimulated to action until sex hormones begin to be 
secreted, and lower its threshold to stimuli. 

Conversely the whole train of events constituting adolescence is 
initiated by the hypothalamus (or by higher centres in the brain). The 
time of the beginning of adolescence seems to depend on the brain 
reaching a certain stage of maturation. Possibly the hypothalamic 
centre responsible is the very last part of the brain to become mature 
and functional. The postponement of adolescence is seen only in the 
primates, and seems to be an evolutionary point of some importance 
(Tanner, 1953). 

You know, of course, that sometimes pathologically early adoles- 
cence occurs in girls and boys, and when this happens the events of 


adolescence proceed all the way to spermatogenesis in the male and, 
if the circumstances are favourable, to pregnancy in the female. This 
raises another very interesting problem. The hormones of the anterior 
pituitary are present in the anterior pituitary from birth onwards, as 
far as we know; certainly they are present very early. They are not 
normally released until a certain time, probably controlled funda- 
mentally by genetical factors, and proximately by the maturation of 
the hypothalamus. Tumours in the hypothalamus may cause release 
of anterior pituitary hormones, which bring about the secretion of 
testosterone or oestrogen, as the case may be, and the changes of 
puberty, including spermatogenesis and the production of ova. There 
are two exceedingly interesting papers about this; one is by Gesell 
and others (Gesell, Thoms, Hartman and Thompson, 1939) and 
another is hidden in a journal called the Reports of the Royal Berk- 
shire Hospital (Le Marquand and Russell, 1934). Le Marquand's 
case concerned a boy of two years old. He did not quite go to sperma- 
togenesis, but he had adult-sized genitalia and had erections. He did 
not have any seminal emissions. Presumably there wasn't any semen 
to emit, but he had the sex-drive quantitatively appropriate to the 
adult. On the other hand, he behaved in a way which was more 
appropriate, both from an analytical point of view, and as far as 
common sense would have it, to the two-year-old child. He would 
attempt to rub his penis up against women's legs, meanwhile sucking 
his thumb. He had no idea about adult sex behaviour, but presumably 
he had his nervous reproductive centres stimulated by the hormone. 
Evidently the hormones had got out of step with his brain, and there 
must have been further maturation processes due to occur in the 
brain before he could produce adult sex behaviour.* In the Gesell 
case a similar divergence between mental and reproductive matura- 
tion occurred in a girl aged about four when she began to menstruate. 
Her brain, of course, was nearer to adult concepts. The whole 
business of adolescence is highly instructive, because we know much 
more about adolescence on the physical and physiological side than 
about any other period of childhood. Questions about the hormone 
secretion and growth at that period really can be answered now to 
some extent, whereas questions about the earlier periods cannot be 
at all. 

Now lastly, I would like to draw your attention to a short hst of 
some questions and of data required for answering them; then try 
finally to give you some picture of the growth process. 

* This child was also said to have shown quite abnormally precocious interest 
in and skill at manipulating mechanical things such as farm tractors, motor cars, 
and motor cycles. 


Some Questions 

(i) Is advancement in physical development as judged by bone 
maturity and other means related to the age at which children reach 
the psychological milestones ? 

(ii) Hormonal and internal environments differ fairly consistently 
from one child to another; is early psychological development related 
to these differences ? 

(iii) Endomorphic children appear to have earlier adolescence than 
endopenic ones ; do they also have an earlier psychological develop- 

(iv) Girls are ahead of boys in physical development up to puberty. 
Are they ahead also in psychological development ? Do both effects 
disappear when (iii) is taken into account ? 

(v) What is the relation between finalization of E.E.G. pattern and 
time at which the adolescent growth spurt starts ? 

Some Data Required 

To settle some of these and numerous other questions we therefore 

{a) descriptions of the growth of the brain in the human. Such 
studies should give cell counts of the various parts and nuclei, 
allowing inferences as to when anatomical development of each 
ceases. They should also give details of myelinization and of any 
other histochemical criteria allowing inferences as to the state of 
functional development; 

{b) longitudinal studies of endocrine and biochemical development 
in association with highly accurate measurement of physical growth 
in a group of children ; 

(c) both {a) and {b) in apes, with facilities for experimental altera- 
tions of both brain function and internal environment. 


The growth process seems to me best visualized as a series of waves 
of activity. There is a fundamental ground plan to this which consists 
of a steadily decreasing velocity of growth from early intra-uterine 
hfe onwards, and on it is superimposed at least one wave of increased 
velocity at adolescence, and possibly another. The waves of growth 
do not hit each part at the same time. First the brain is growing 
faster and then the chest is growing faster and then the length of the 
legs is growing faster. We can see the flowing of that wave of activa- 
tion, or whatever one likes to call it, very well in the sequence of the 
epiphysial unions. This is a very constant sequence, and it is not 
much affected by whether everything is occurring relatively early or 


relatively late. Girls are ahead of boys in physical development at 
birth and they remain ahead all the way through. Their epiphyses 
join in the same sequence as in the boys but always earlier, and 
their adolescence is about two years earlier. 

The only thing to which I can liken this whole process is a series 
of clocks which are wound up and gradually run down. Now, of 
course, you can interfere with the running down of them by sticking 
your fingers in the way, which is what happens if you malnourish the 
child. But let us forget about that for the moment and get to the 
fundamental things, presuming that everything is all right for the 
child. The clocks run down one after the other and as one runs down 
it seems to initiate the running down of the next one, and so on. Con- 
sider adolescence — something in the mid-brain reaches a particular 
point of development, and at that point impulses go for the first time 
to the pituitary, and the next clock, the pubescent endocrine one, 
begins at this time to run down and drags the child through that 
phase. When we think of phases of development in this way it seems 
to me quite irrelevant to talk about critical phases as though there are 
times when an organism is particularly susceptible or particularly in 
danger of some influence. I do not think there is at present any 
evidence in physical or physiological growth for sharply defined 
critical periods in man, but there is, I think, a probabihty that an out- 
side influence can make itself felt only between two points of time in 
the growth pattern. Searching my mind for an example of this, I can 
only find a very bad one. It is a natural one, but you will see why it is 
bad. If you give follicle-stimulating hormone of the anterior pituitary 
to a child before a certain stage of intra-uterine development it has no 
effect, because the ovaries cannot secrete oestrogen. If you give it after 
the menopause (that is, after rather a long period of growth has gone 
by), it has no effect either. If you give it in between, then it produces 
oestrogen secretion and all the things that go with that, including, of 
course, behaviour changes. One can conceive of susceptible periods 
like this occurring in the growth of the child very easily. I don't 
think, in fact, that there is any evidence that such things are there, 
but that may well be merely because we haven't got the evidence. 
Dr. Lorenz's experiments on mother-following in the goose and 
others are certainly evidence that at particular times certain neurons 
(if I may change languages) are capable of having their usual con- 
nexions made in a different way, though, as far as I am aware, no 
cast-iron evidence on the physical and physiological side exists about 

That is all I want to say to start with. May I repeat that I feel that 
probably my best use in this group is to answer such questions as 
you wish to put on basic data from the physical side. 



I should like to ask a question about critical periods, and to give 
an example how it is in birds. There is something like the critical 
period, something like the finger that is stuck into the clock which is 
running down. There are periods when certain organs are growing 
fast and must grow fast, and evidently cannot stop growing. If you 
take a young goose in the middle stage of growth and malnourish it, 
it just grows much more slowly; then if you feed it up again it simply 
grows fast again and develops into a normal bird. Now in water- 
fowl, the wings grow quite slowly, nearly not at all, until shortly 
before fledging, and then suddenly there is a tremendous spurt, they 
catch up with the rest of the bird. Now if, during that period, you 
subject the animal to malnutrition, then the wing goes on growing as 
if nothing had happened to the bird, the wing is hardly affected, but 
the bird gets thinner and thinner and usually gets tuberculosis and 
dies. I would hke to ask if there is something similar in the case of 
children, for example, towards puberty. 


I would rather doubt that. I think the evidence is at least very 
equivocal. I am most interested by what you have to say, and I would 
be even more so to discover by what mechanism, I suppose endocrine, 
the bird gets its nitrogen and so forth out of itself and into its wings. 
Of your first example there is evidence in man ; the child will starve 
and be generally retarded and then come back to normal later on 
when he is well nourished once again. 


In pregnancy I think there is an equivalent. I think that the foetus 
does get the nourishment even in malnutrition, and I think also that 
in breast feeding to a considerable extent the milk continues at the 
expense of other portions of the body. 


But the foetus gets smaller. 


Yes, but there is a tendency for the foetus to be preserved at the 
expense of the mother. 


I have a better example. There is some such relation between the 
important and the unimportant feathers on birds. For instance, 


certain feathers must be kept for one year and some others are shed 
within a few weeks after fledging. If malnutrition causes a shortage 
of horn-producing substance the horn that is available goes into the 
primaries and the primaries are the last feathers to be affected by 
malnutrition. Some other feathers may be so dystrophied as to be 
hardly present at all. I have some sheldrakes at home that had to 
economize on horn substance, but in all of them the primaries are 
beautiful. Cold-blooded vertebrates can grow as slowly as they 
please. They do not mind malnutrition at all. A newt can catch up at 
any period of its life; when you start feeding it again it will grow. 
The bird cannot do this. 


I want to put the general zoological proposition that those char- 
acters which are phylogenetically the most recently acquired are the 
ones which ontogenetically are the most favoured in times of stress 
or emergency. As Dr. Tanner told us, the data on brain growth are 
extremely sparse and ill-favoured ; but in the few measurements that 
have been made of the weight, sizes, and general proportions of 
different parts of the brain when people have been subjected to 
malnutrition, inanition, and stress, such as in the concentration camps 
in Germany, those parts of the brain which have retained most of 
their normal structure were those which developed last: that is the 
association, temporal, and frontal regions, while the primary motor 
and sensory regions tended to be relatively under-nourished and 
dystrophic (Wulff, personal communication). 


I would hke to raise a question which might underhne a good deal 
of the discussion. I am thinking of the concept of innateness. It seems 
to me there is possibly an implicit idea that the environment can be 
passive and that the organism then just unfolds actively in it. Now I 
do not know whether anybody means that, but I would at least like 
to put into words the opposite point of view that there is a constant 
interaction and that all behaviour of which growth is a manifestation 
is a result of interaction between organism and environment with an 
intricate interplay in a reverberatory process, but there is no time 
when the environment is inoperative. Therefore, if we endeavoured 
to change the environment in a pertinent way, we would be able to 
modify any form of so-called innate or unfolding behaviour. When 
we do not see this it is only because the environment remains constant 
for the particular manifestation of growth or behaviour or unfolding. 



Well, I should consider that any genetically determined character 
is determined only in respect to a certain range of modifiability. If 
you take any instinctive action of a bird you can only decrease the 
intensity along a quite definite gradation down to zero; you can't do 
anything else. You would not get the shghtest actual qualitative 
variation of his instinctive activity. If you take any more highly 
differentiated character, let us say, the structure of a bird's feather, 
its form cannot be changed as can the leaves of certain plants. I 
think the question of whether a character is modifiable, and to what 
extent it is modifiable, to what extent it is interactive with environ- 
ment or whether it may be totally independent, endogenous, is a 
question which can only be answered by an experiment for each 
individual case. You cannot say out of hand that all is dependent on 
environment or that all is independent of environment. 


The point I am trying to make is: is it on theoretical grounds 
tenable to assume that no conceivable change of blood chemistry, 
irradiation, operative procedure can change this particular character? 
It does seem to me hard to accept that theoretically. It seems to me 
that manifestations of behaviour are always interactions of the 
organism with the environment, because there is no such thing as 
organism without environment. 


In theory, of course, no genetic mechanism can operate in vacuo ; 
if the animal is completely starved, there is not much environment, 
and there is very soon no animal. Nevertheless, particular characters 
— I entirely agree with what Dr. Lorenz says — vary from being not 
at all dependent on the environment, except that if there is a lack of 
a particular amino-acid, etc., they just won't develop at all and you 
get a thoroughly pathological beast, to the other extreme of being 
genetically scarcely controlled at all. I feel that the placing in opposi- 
tion of environmental and genetic features in a general sort of way is 
quite misconceived, because each problem is a quantitative and a 
particular one, and is neither qualitative nor general in any sense. 

This, I think, brings us to a matter of importance for future research. 
The question arises as to how much the human skeleton, for example, 
can be modified during growth by various environmental factors. We 
don't know the answer to this, and as Dr. Lorenz has imphed, and as 
I fully agree, it will not be the same for all parts of the skeleton. 
About the only thing that we do know is that you can retard puberty. 


At the behavioural level this discussion leads on to things which 
seem to me intensely important, because, if I understand Dr. Bowlby 
rightly, what he says about the effects on children who were separated 
from their mothers at particular periods of growth is that environ- 
mental factors have irreversibly changed behaviour. In neurological 
language this implies that environmental factors have produced 
altered formations of neurological connexions in the brain and/or 
syntheses of different proteins in the neurons. To my mind one — 
though by no means the only — way in which Dr. Bowlby's thesis 
could be validated, would be by experiments to determine that these 
physical alterations have in fact occurred. There we have got to 
rely, I suppose, on the electroencephalographers, because there is 
nothing much else we can do with the brain except take electric 
currents off it. You can investigate this with animals, though. 


I wanted to bring up another point and to link it up with one that 
Dr. Fremont-Smith has made. I do not feel quite happy about some 
of the implications of Conel's ideas on the cerebral growth of the 
brain. I am referring to the remark that the first centres to develop 
are the primary motor, primary sensory, primary visual, and primary 
auditory, and that the cingulum and those structures generally 
referred to as emotional develop later. It seems to me that we ought 
to make a distinction between what we might call morphological 
growth and functional growth, especially with reference to myeliniza- 
tion, because of course it is quite true, cytologically speaking, those 
centres are the first to develop, but from the point of view of myelini- 
zation of pathways quite different centres have precedence. The 
vestibular structures, for example, are fully myelinized at the age of 
four intra-uterine months (Minkowski, 1924, 1925). From the point 
of view of myelinization (see Krapf, 1950) not only the vestibular 
but also the cingulum and the olfactory structures seem to develop 
before those others. It would seem that the functional growth as 
witnessed by myelinization has primary importance, and this is where 
I think the subject hnks up with Dr. Fremont-Smith's point, because 
according to my information it seems that the use of structures is 
conducive to a higher speed in myelinization, which would, of course, 
mean that environmental pressure, which causes certain structures to 
be used at a higher rate, would influence maturation. 


I am very interested in what you have to say. As I understand you, 
the use of a tract causes myelinization or accelerates myehnization. I 


would very much like to know what the evidence for this is, and 
whether the neurologists here think it is true or not. 


No, this is not true. There is no evidence that the use of an organ, 
or of the nervous system, accelerates growth. 


Can you retard it by the deprivation of experience ? 


Nothing is known about that, as far as I'm aware. 


I think that there is a behaviouristic parallel though. Grohmann 
(1939), of Vienna, assessed by a very nice quantitative method the 
maturation curve of flight. He did this by putting standard perches at 
definite distances from the pigeon-cote, and then he standardized the 
distance the young pigeon would fly after fledging. Then he got a curve 
for the distance at stated times after fledging. He took always two 
pigeons of one brood, and put them in longitudinal boxes where they 
could not open their wings (this is not anti-genetical, because the 
pigeon is a cave breeder, and breeds in cavities too small for the 
young birds to get exercise). Now one bird of a brood was put in a 
separate box, the other was left free, and then the boxed bird was 
liberated two days after the fledging of the normal one, and so on. 
Now the result was this : the imprisoned bird jumped with a bounce 
up to the normal curve and then went on following the normal curve. 
This happened to the extent that the bird which was liberated after its 
brother had reached the end-point of the curve, which is circling in the 
air, came out of the box and circled immediately, and a bird kept cap- 
tive much longer got into a frenzy of flying after being liberated and 
then had such a hangover that it wounded its wings and could not fly 
for weeks. It had muscular atrophy, and it is very interesting that mus- 
cular atrophy always sets in at a time when a bird would normally 
move its wings. A bird does not get inactivity atrophy as long as it is 
in the nest. Then Grohmann did the opposite experiment; he took 
the control bird to its perch and made it beat its wings. He took it 
out, put it on his finger, quickly lowered the perch — the bird would 
flutter. He subjected these experimental birds to a standard time of 
fluttering every day. The result was that their normal flight develop- 
ment was retarded in proportion to the forced flying movement they 


did. From this I do not conclude that such a thing as acceleration of 
maturation is impossible. I am quite sure that it will be possible. 
But I think this story illustrates what Dr. Grey Walter emphasized 


There might, of course, be some difference between what is possible 
in already fairly developed individuals and what is possible in a still 
quite undeveloped individual. In this context, I would like to refer 
again to Minkowski's work (1924, 1925). According to his studies 
there is apparently a relationship at least in the foetus between use 
and rate of myelinization, but I am quite prepared to accept that 
later on this does not take place at the same rate and quite possibly 
does not take place at all. 


Even for the muscular system, which one might suppose would be 
an easier system to study, there is little real evidence that exercise 
will cause anything but a temporary hypertrophy of muscle fibres 
which disappears again once the exercise is stopped (Tanner, 1952). 
That is certainly true in the adult, and nobody knows, as far as I am 
aware, whether it is any different in the child, because nobody has 
ever done any experiments which stand up to criticism in this regard. 
It should certainly not be uncritically assumed that taking a child and 
making it do exercises for a while is going to have any effect at all, 
except to make the fibres somewhat bigger just while the exercising 


Dr. Tanner, I was surprised that you did not say anything to us 
about the individual differences in constitutional types. In connexion 
with this critical period, is it not possible that periods that are vulner- 
able may be points of refreshment with slowing down or speeding up 
of a variety of these systems which are growing at different rates ? 
If we have material on children of diff"erent constitutional types and 
we could extrapolate points in the pattern of differential growth 
curves, then we might get a more abstract statement of criticalness. 


I hope you will raise that point later on because I should hke later 
to say something about constitutional differences. There is also 


one point of importance about a difficulty which fundamentally per- 
plexes the whole field, at least on the physical side. We study the 
height of the child — I have shown you some curves ; they are very 
regular. We study parts of the height measurement and find that at 
one stage the neck is growing very fast, at another the middle of the 
trunk is growing very fast, a little later the legs are growing very fast, 
and so on. Now bring it down to smaller elements. These, of course, 
we cannot study, so that we cannot really determine whether our 
regular curves are made up of sudden tiny, localized spurts. As far 
as the curve of stature is concerned I doubt whether it is, but if we 
think of the growth of the brain, in the foetus or at other times, then 
we get a little bit nearer the point, because it may be there that 
you get a little bit growing fast and all the rest staying still. I am 
not saying this does happen, but if it did we could not at present 
detect it. 

The other thing that perplexes the field is that for the development 
of the brain we must have longitudinal data because of differences 
between individuals. Now you cannot have longitudinal data on the 
histo-chemistry of the brain, because once you have done the histo- 
chemistry of the brain of a child, the child no longer exists, so what 
are we going to do ? In animals we can create to some extent identical 
individuals by breeding isogenic strains of animals and bringing them 
all up in the same environment. The only trouble with this approach 
is that the processes that are beautifully demonstrated in some species 
may not be present in others, particularly man. For example, the 
whole sequence and timing and effects of puberty in the human is 
vastly different from puberty in rats. The chimpanzee is two-thirds of 
the way to the human from the rat in this regard, and the only thing, 
it seems to me, is to get hold of a lot of chimpanzees to get as near the 
human as possible, and I know this is very expensive. 


I was going to ask Dr. Tanner exactly this question about the scale 
of measurements: what is the scale on which one plots results? 
Clearly, if you look at the mitotic figures in the individual cells, you 
see ample and classical evidence of the abruptness of growth. The 
cell remains very much the same for a long time and suddenly, in the 
space of hours, the mitotic spindles part and the whole thing almost 
explodes into two cells, so that, at some scale, one does find very 
sudden changes and abrupt development of functions. In the case of a 
relatively simple process, such as mitotic division in the normal 
growth of the bodily cells, one would not expect to find anything very 
abrupt, because the mechanism is statistically homogeneous. In other 


words, you have a large number of muscle, bone, skin, and so on, 
cells, all doing very much the same sort of thing and distributed in 
homogeneous physical patterns. When, however, you come to study 
mechanisms of adaptation and behaviour, such as Dr. Lorenz studied 
and such as we study in the brain, there you are studying a relatively 
intricate mechanism made up of non-homogeneous statistical com- 
ponents, and in such assembhes you do find evidence of abrupt 
change; even a motor-car either does go or does not go, although 
the different parts of it may be statistically smooth or stationary. I 
think this question of the scale on which things are studied is a very 
important one for us to consider. 


I would like to ask a very elementary question about the develop- 
ment of the central nervous system. Is it probable that on the whole 
the lower centres develop before the higher centres — for instance, is 
it probable that the mid-brain is functional before the cortex, or 
don't we know even that ? 


I think we don't know, and I think probably, if you will excuse 
my saying so, you are lumping together too much when you ask the 
question that way. Some parts of the mid-brain must function quite 
early but, as I said before, the part of the mid-brain somewhere near 
the tuber cinereum which functions to kick off the adolescence 
mechanism doesn't mature till very late. I want to ask Dr. Grey 
Walter, is this part of the brain the last one of all to mature ? It would 
be very interesting if what happened was that human reproduction 
was held up until the brain was really completely matured. The 
question arises as to why we should have this very late puberty any- 
way. I presume the evolutionary answer is that before we breed we 
want to be as bright as possible in order the ensure the survival of 
the child. 

GREY m^alter: 

I quite agree with you that the idea of 'levels' of mid-brain and so 
on is really much too ingenuous. 


I think that the work of W. R. Hess is extremely interesting in 
regard to what happens in the mid-brain. We know from Hess's 


experiments that very persistent and highly differentiated motor 
mechanisms, behaviour patterns, are localized there. These may be 
activated by hormones dependent on gland maturation and at the 
same time be dependent on the nervous system. You can induce 
mother behaviour in a common barnyard cock by injecting it vi'ith 
prolactin because he has the nervous mechanism for brooding young. 
But you cannot do it in the White Leghorn, because in the central 
nervous system of the White Leghorn brooding has evidently 
suffered a defect mutation ; brooding cannot be produced either in the 
hen or in the cock. A similar question is interesting in regard to the 
behaviour of your true pubertas praecox, and I should be very 
interested in having details of their behaviour, whether they were 
girl-conscious, whether they courted beautiful girls, and so on. 


I am afraid I cannot answer that very well. It is about a couple of 
years since I read one of the papers. All I can remember is that the 
little boy of two years old showed a marked preference for some 
women over others. 


Were they pretty ? 


No, it was a more basic phenomenon, if you will pardon my calhng 
it so. It was said in this paper that he did not pursue women who had 
passed the menopause, but he did become very difficult in the street 
because if he saw someone under forty in a nice pair of silk stockings 
he would make a dash for them and rub himself against them, and 
this got quite embarrassing. But he did not go at all for young, 
prepubescent girls. Neither did he ever make sexual advances towards 
his mother. This patient was in the Royal Berkshire Hospital, and I 
may add that he was studied only by ordinary doctors with both the 
advantages and disadvantages that that implies. There were no 
psychologists around, so that the study is not influenced by precon- 
ceptions of a psychological nature. 


We had a prepubertal case, a child of six who showed very good 
taste with the nurses. He knew the pretty nurses extremely well. 



Well, that shows the difference between two and six, which is just 
what we are after. 


I should like to put three questions to Dr. Tanner. The first is as 
follows : is the present state of our knowledge of allometry and of 
non-parallel growth rates sufficient to enable us to understand the 
critical phases and stages of growth? In other words, despite the fact 
that there is a slow, continuous progression, as Dr. Tanner reminded 
us, are there for the human species rearrangements in the balance 
of the constituents and allometries which enable us to understand 
or at least to guess the explanation of the stages of growth, and to 
describe the critical phases ? For us psychologists it would be impor- 
tant to see if any co-ordination is possible between critical phases 
found by developmental biology and critical phases which we 
observe in psychology. 

My second question is much more hmited. What is your opinion 
of the frequent statement that the age of sexual maturity in girls is 
earlier than in boys, since the appearance of menstruation is earher 
than that of spermatozoa ? Can we consider that as an estabhshed 
fact ? Much research over the last forty years has tended to prove the 
contrary. Is this current belief not due simply to the fact that girls 
start on the phase of accelerated growth before boys ? This is impor- 
tant for the understanding of adolescence on the psychological level. 

My third question deals with the lowering of the age of puberty, 
which seems to have been occurring during the last fifty years, that is 
since it has been possible to obtain rehable data. Is the variation in 
age of puberty with different social environments considered by 
Dr. Tanner to be a well established fact, explainable by nutritional 
factors ? In other words, what connexion can we estabhsh, in a general 
way, between the evolution of the organism and environmental 
influences, 'environment' being very widely interpreted ? 


I will take the questions in the order in which they were asked. As I 
understand it, your first question implied that the study of allometry 
in the physical side of development gives some background for the 
suggestion that there are critical phases during growth. When allo- 
metry was first thought of by Huxley and Teissier and others, they 
imagined that by plotting the growth of one dimension against 
another, both in logarithms, they produced a curve with a sudden 


sharp change of direction in it. Later work has shown this notion to 
be quite fallacious, at least for mammals. The sharp change of direc- 
tion resulted only from drawing various straight hnes through points 
which really show a gradual and continuous curvature (Tanner, 
1951). Any idea that critical phases may exist because of that formula- 
tion of allometry is ill-founded and incorrect. That answers the 
question without becoming technical, which I am very wilhng to do 
if we have the time. 

Now the second question — the difference in age of sexual maturity 
between girls and boys — first, is it established that girls begin puberty 
earlier than boys ? Yes. Definitely, clearly, no question. About two 
years on average. It varies between individuals of course. There are 
some girls who enter puberty after some boys. There are some inter- 
esting data that suggests that those girls who develop late become 
relatively more masculine in physique and those boys who develop 
early become relatively more feminine (Bayley, 1943). I don't think 
this is entirely estabUshed, but it is reasonable, and it is a subject 
which would stand looking into. So much for the spurt in physical 
growth. However, if I understood Dr. Zazzo correctly, he is more 
concerned with the attainment of actual reproductive maturity in the 
strict biological sense, that is, spermatogenesis, copulatory behaviour, 
the ability to produce children. The evidence here is not so complete. 
I would imagine girls are capable of breeding earlier than boys : on 
the other hand, there is a well-documented period of adolescent 
sterihty. The course of puberty is this : there is an increased velocity 
of growth in stature and in all the physical dimensions, and shortly 
after the peak velocity has been reached and the speed is slackening 
to zero, menstruation begins (see Fig. 5). This relationship is remark- 
ably constant. The uterus grows fairly late and gradually becomes 
functional, and as growth in height begins to finish, menstruation 
begins. Now, despite the uterus getting to the stage of bleeding, the 
early cycles, which are often not so regular as the later ones, may be 
anovulatory, and due only to oestrogen withdrawal, and this may go 
on for perhaps a year, eighteen months, two years. There is some 
physiological evidence for this and also a good deal of anthropo- 
logical evidence that repeated intercourse at that time does not lead 
to pregnancy. It may well be that there is a similar period in boys. 
There is no evidence about this as far as I know, but it seems a 
reasonable thing to assume. That's as far as I can go towards 
answering your question on the facts we have at present. 

Now for the third question : is the age of puberty related to environ- 
mental factors and social class? Yes, it can be related to environ- 
mental factors; for instance, adolescence was retarded in parts of 
Europe where the food supply was diminished by the war. There is no 


FIG. 5 



















. MENARCHE 10-12 

* * MENARCHE 12-13 


_J I \ I 

I I I 

7 8 9 10 II 12 13 14 15 16 17 

After Tanner (1953); data from Simmons and Greulich (1943) 

question that malnutrition can postpone it. I am not sure whether 
there is any literature which distinguishes clearly between different 
social classes and whether any class effect could be related to things 
other than malnutrition, or whether indeed it could be related to 
genetics. As to other environmental factors such as climate, it has 
often been said that puberty is earher in the tropics. But the data on 
which that statement was based are very poor. There has been one 
good study by Ellis (1950) recently, which showed that puberty 
occurred at very much the same time in Nigeria as in Western 


Europe. It was shown in California that different racial groups had 
menarche at different times, but this may be chiefly due to genetical 
factors (Ito, 1942). The differences were not in any case very large. 
So this question cannot be answered certainly, but short of malnutri- 
tion it seems that if there is any social class effect, it isn't a big one. 


Would you include the work that Peller (1940) did on the effects 
of starvation in the mother during pregnancy on delayed menstrua- 
tion in the daughters ? 

I don't know that work ; I would like to have the reference. 


These studies were based on Viennese girls whose mothers were 
pregnant with them during the worst period of starvation in Vienna 
in World War I, and there was a consistent postponement of the 
age of menstruation in the daughters. The implication was that in 
handling first-generation data on ethnic groups we perhaps ought to 
go back a generation in order to trace the causes of any differences. 


Or rather to the foetus. I regard this with a certain amount of 
doubtfulness because most of the data say that semi-starved foetuses 
such as we are assuming to exist would only be small and perhaps 
developmentally retarded at birth. Most of the data (though these 
data are not cast-iron stuff at all) would rather indicate that if the 
children were later well nourished they would catch up on their 
growth curves rapidly, and certainly not be thrown back all that 
long time during growth, as you imply. 


There were other concomitants also. There were pelvic distortions, 
sometimes prolonged amenorrhoea after puberty, and diflBiculty in 

carothers : 

It has been commonly written and commonly assumed in the past 
that puberty was early in the tropics, but this was largely based on 
estimates of age. The African never knows his age and these estimates 


were grossly faulty. They are not quite so faulty at the age of puberty 
but at later ages it is easy to be ten years out, and the age is nearly 
always under-estimated. 


With regard to the matter of the onset of menarche in South 
America, what I have been able to find out from gynaecologists' 
reports about the matter is against any differences between the 
temperate zones and the tropical zones. 



The Behaviour of New-born AnencephaHcs 
with various Degrees of Anencephaly 


We have made a prolonged study of the scheme of integration of the 
motor functions by the central nervous system. Various methods can 
be used. The best way is to study the development of the motor paths 
in the embryo, foetus, and newborn, a method initiated by Monakov 
and Minkowski at Ziirich and experimentally perfected by Windle 
at Chicago. Phylogeny and ontogeny supply us with concordant 
data which enable us to reconstitute the scheme of integration 
(Auf bauplan) of the motor functions provided that these data are 
analysed and interpreted according to functional criteria (Monnier, 
1946). If we admit as a criterion of integration the synthesis of the 
elementary mechanisms in a function adapted to an aim, we can say 
that the scheme of integration advances by stages at well defined 
times. We have distinguished the following stages : 

1. Integration of motor functions in respiration and nutrition, 
functions of mime and vocal expression, protective functions with 
predominance of flexion mechanisms and functions of prehension 
(end of foetal life, and birth) ; 

2. Functions of active orientation of head and eyes (two to three 
months) ; 

3. Functions of lifting the head (two to three months), the trunk 
(five to six months), the legs, retaining position (seven to ten months); 

4. Functions of progression, locomotion, and regulation of equih- 
brium (eleven to fourteen months) ; 

5. Articulate language (fifteen to twenty-four months); 

6. Technical manual dexterity characteristic of working man 

Although it is always possible to study experimentally in the animal 
the correlations between the stages of development of motor functions 
and the stages of differentiation of anatomical functions, which give 


us information on the integrating function of the differentiated 
mechanisms, the same cannot be done with man. This is why the 
newborn anencephahcs, with their rudimentary bulbo-spinal, ponto- 
bulbo-spinal or meso-ponto-bulbo-spinal brains afford us an excep- 
tional opportunity of defining the correlation between a stage in the 
organization of motor functions and the corresponding stage in the 
morphological development of the nervous system. The film will show 
the motor paths of ponto-bulbo-spinal anencephahcs (rhomben- 
cephahc anencephaly), then those of a meso-rhombencephalic 
anencephahc and finally those of Camper's anencephahc with a 
well developed mesencephalon. At the same time anatomical sections 
of the brain stem will be shown, illustrating the degree of develop- 
ment of the nervous integrating mechanisms. 

(a) Rhomb encephalic Anencephalus 

We have been able to observe four anencephahcs whose brain was 
hmited to the pons, medulla oblongata, and cord. In such cases the 
vegetative functions are very poorly regulated. The blood supply to 
the peripheral area is insufficient and respiration is irregular. Inges- 
tion of food is badly co-ordinated, especially sucking. The tempera- 
ture is very labile (poikilothermy) and the infant can hardly survive 
more than one or two days. There was no spontaneous activity and 
no periodicity of states of wakefulness and sleep. Motor activity con- 
sists mainly of defence behaviour by flexion with intense protective 
reflexes to noxious tactile stimulation or to acoustic or vestibular 
stimulation. Integration at this stage is characterized by poor localiza- 
tion of reflex responses, a tendency to irradiation, bilateralism, and 
even generalization of response (mass reflex). Stimulation of the sole 
of the foot, the malleolus, or the Achilles tendon produces, not a 
Babinski phenomenon confined to the big toe, but a triple retraction 
of the leg on the side stimulated, often also of the other side, and 
sometimes even a reaction of the upper limbs accompanied by clonic 
trembling of hands and arms. Reflexes of flexion posture and prehen- 
sion reactions are highly developed. Rudimentary balancing reactions 
are observed with extension of the arms and flexion of the head, as in 
Moro's reflex. Although conditions for development of mental 
activity do not perhaps exist it can at least be said that mechanisms 
for protection against noxious stimuli and prehension to counteract 
the dangerous effect of weight are accompanied by expression 
phenomena capable of interpreting certain rudimentary affective 
states. The mechanisms responsible for these primordial integrations 
are the reticular formations of the medulla oblongata and the pons 
(Tegmentum pontis), with the posterior cords and the cranial nerves 
V to XII (MoNNiER and Willi, 1947). 


FIG. 6 


Norma! ; Anencephalus 

1. Rhombencephalic Anencephalus 

Monnier & Willi 1947 

2. Meso-rhombencephalic Anencephalus 

Monnier & Willi 1953 

3. Mesencephalic Anencephalus 

Gamper 1926 

{b) Meso-rhombencephalic AnencephaUc 

I had the opportunity of observing with H. WilH an anencephalic 
whose brain was Hmited to the caudal area of the mesencephalon 
(isthmus) together with the pons, medulla oblongata, and cord. It 
was possible to keep the newborn alive for fifty-seven days. It showed 
good circulation, irregular and sometimes periodic respiration, 
poikilothermy and mediocre regulation of blood sugar. Spontaneous 
activity was limited to a few movements of the face and lips. The 
infant lay on one side with arms and legs flexed. It had the advantage 
over the type previously described of having a more physiological 
posture during sleep. The hands were prone and legs supine. It slept 
most of the time and showed little movement — only lazily or by fits 
and starts — when stimulated. The head sometimes moved slowly 
from one side to the other. 

The somatic-motor integrated functions consisted mainly of defen- 
sive flexion reflexes released by exteroceptive tactile, thermic or 
chemical stimuh, more or less noxious. Very intense stimuh provoked 
a generalized defence reflex (mass reflex). Grasping mechanisms were 
well developed in both hands and feet ; they could be clearly observed 


during trophotropic activities, such as sleep and feeding. Functions of 
mime and vocal expression were well developed : contortions of the 
trunk, vermicular movements of the head and limbs, sometimes cer- 
tain facial expressions of affective pleasure reactions with stretching, 
extension of the arms, yawning and sighing. Passive extension of the 
leg induced, for example, a Moro reflex, with deflection of the head, 
extension of the arms, especially the right arm, opening of the mouth, 
and sighing. As to vocal reactions, they were limited to a sort of 
hiccough and stridor. The functions of extension and of hfting against 
weight remained rudimentary : reactions of extension and abduction 
of the arms (stage 1 of Moro's reflex) with lifting of the trunk. As 
regards functions of active orientation of the head in space, they were 
limited to mechanisms of buccal prehension (prehension of fingers or 
bottle with lips, lifting of the head under the influence of propriocep- 
tive excitation from the stretched nape of the neck or from the vesti- 
bular apparatus). The lifting of the trunk in sitting position by 
passive extension of the legs can also be considered an elementary 
lifting mechanism. 

In this type of anencephalic we were able to observe in the course of 
weeks the transformation of an inadequate reaction (turning of the 
head towards a noxious stimulus : prick of a needle) into an adequate 
flight reaction. This adequate reaction appeared seven weeks after 
birth ; it was at times so intense that it even took the place of buccal 
prehension behaviour; the mouth moved away from the finger 
presented instead of seizing it (Monnier and Willi, 1953). 

(c) Mesencephalic Anencephalic 

In the anencephalic of pure mesencephalic type, described by 
Gamper (1926), all the vegetative functions are well regulated, 
especially respiration and circulation. There is a distinct alternation 
between periods of sleep and waking activity. General spontaneous 
activity is more intense (movements of the tongue, contortions). 
Locomotor paths are highly developed ; they consist mainly of reflexes 
of crossed extension, automatically released, and very clear successive 
induction phenomena in sitting position (epileptoid trembhng). The 
functions of facial and vocal expression are better elaborated : smiles 
and tears. The Moro reaction, induced by a puff of cold air, is shown 
again by extension of the arms, but also by deflexion of the head, 
with a yawning reaction, whereas in the rhombencephalic anen- 
cephalics we had mostly a head-flexion reaction. The functions of 
protection and defence are well developed. Extension and lifting 
functions are definitely better developed than in the preceding 
anencephalic types. The same is true of the tonic neck reflexes, the 
reflexes of lifting of head and trunk, and the reactions to vestibular 


stimulation of rotatory or linear acceleration. This is also the case 
with all the functions of orientation of the head and upper body in 
space. The paths of oral prehension are so well developed that they 
produce in the anencephaUc an effect of an automaton, which always 
directs its mouth towards the finger that touches it. Here too one can 
say that the development and maturation are characterized by greater 
spontaneous activity, mainly pre-locomotor, by better diiBferentiated 
functions of facial expression, giving the impression of affective 
reactions, better elaborated functions of lifting and motor orientation 
and especially by a closer and more adequate adaptation of reactions 
to objective (Gamper, 1926). 

Dr. Monnier then showed a film illustrating these descriptions. 


I would like now to discuss the question of reactive patterns 
changing with the development of the nervous system. My film 
showed a meso-rhombencephalic anencephalus turning the head 
toward a source of nocive stimulation (prick of a needle) during the 
first six weeks of life, and then turning the head away from the 
stimulus. This latter protective reaction could be ehcited chiefly from 
the seventh week after birth. Ipsiversive and contraversive patterns of 
head turning were probably both integrated in the brain at birth, but 
the threshold of the two systems was different : low threshold for 
ipsiversive turning during the first weeks (tegmental reaction), higher 
threshold for contraversive turning (protective reaction). The thres- 
hold of the latter decreases only seven weeks after birth, as a sign of 
greater maturation of the brain. 


What is the reaction in the normal infant ? 

Always away from the stimulus. 


I beg to doubt that. There is a great difierence in different children's 
response to stimuH in different parts of the body. If you study the 
same children with different strengths of stimuH you get very different 
responses. It is a matter of balance between the different muscles, and 
is very often a matter of chance, excluding the effect of well- 
established reflexes. It depends on the precise state of development of 
the motor-groups. Have you studied different strengths of stimuli ? 



When I said : always away from the stimulus, I meant, of course, a 
nocive stimulus ; the prick of a needle. In normal newborns we got an 
ipsiversive turning of the face and mouth on a slight stimulation of 
the lips. 

I should like now to define growth and maturation as a process 
which locates, which circumscribes a reaction, instead of spreading it. 
Diffusion and generalization are expressions of an immature nervous 
system. During growth the reaction becomes more adequate, better 
adapted to its purpose. I gave as an example the meso-rhomben- 
cephalic anencephalus, which showed a better adaptation of the 
head reaction to various stimuli seven weeks after birth ; head turning 
away from the nocive stimulus. In this regard, we have to con- 
sider the notion of threshold. We know in the tegmentum of the 
brain stem there are pre-configurated systems for ipsiversive and 
contraversive head turning. These systems must have at different 
times different thresholds. During the first six weeks, we elicited 
chiefly ipsiversive patterns ; during the second month, as maturation 
of the brain progressed, the contraversive reaction of the head, 
away from the nocive stimulus, became more prominent. There 
was even a kind of rivalry between ipsiversive responses (turning 
of the head with sucking) and contraversive responses (flight 
reaction). It was sometimes even difficult to feed the baby, since 
the flight reaction of the head was too strong, so that the nurse 
could not get near the lips with the bottle. This rivalry was typical 
for the transition state. 

What other conclusions can we draw from our observations and 
from the film, in relation to the problems of psychobiological develop- 
ment ? We may say that in the lowest rhombencephalic type of anen- 
cephaly, the protective patterns of flexion type are predominant. The 
surface of the body is reduced in order to escape nocive stimuli. At a 
higher integrative level (meso-rhombencephahc type) some elementary 
mechanisms of standing and righting postures develop. On the other 
hand, the mime changes with maturation ; it becomes more elaborate, 
sighing appears with contortions of the body, which have sometimes 
an expression of pleasure or displeasure. We do not know, of course, 
if a psychological experience occurs behind these various expressions. 


May I ask Dr. Monnier a question about the conditions during the 
experiments that were shown in the film? I, too, am particularly 
interested in the reversal of the behaviour of the child or creature 
from what one might call a positive to a negative tropism. This 


effect, as you know, can easily be imitated in a model. It requires only 
a simple set of arrangements to exhibit and mimic this particular 
development, but the eflFect produced both in the flesh and in the 
metal depends a great deal on the experience of the individual, 
whether man or machine. For example, in an ordinary healthy baby 
one may see a reversal of attitude or of behaviour, from the positive 
attraction to the breast or the bottle — sucking movements — to a 
withdrawal when the stomach is full either of milk or of wind. I am 
sure most of us with children have suffered from this difiiculty of 
trying to get enough milk into the child; at a certain point there 
comes the avoiding reaction which perhaps may develop in the adult 
into a highly organized negative shaking of the head. It would be 
interesting to know in your film the state of the stomach of the child 
during these two experiments; that is, whether you are quite sure 
that the child is equally well fed or has an equal amount of wind in 
his stomach during both experiments, because one would expect that 
after a certain length of time, varying in the normal child from a few 
weeks to a month, this reversal would occur as a result simply of the 
presence of a distended stomach. That is, the endogenous stimuli in 
•the belly can reverse, in fact, the whole action of the body. 


I am not able to make comments on the influence of the stomach 
and of hunger on the responses of our anencephalic newborns. 


The difficulties of feeding are found in quite normal children. I 
wonder whether you had any evidence of the presence of wind in the 
stomach. Did you X-ray the stomach to see if there was much gas ? 




Isn't your question partly answered. Dr. Grey Walter, in the fact 
that the ipsiversive reaction occurred nearly every time the child 
was tested in the early period, and the contraversive nearly every time 
he was tested in the later period, and therefore it would seem sur- 
prising if there was always an excess of wind in one period and not in 
the other. I might add that I have seen an infant up to the second and 
third, maybe fourth, day having difficulty in getting breast milk, 
being unsuccessful, and finally exhibiting a negative reaction so that 


he couldn't be got to accept the nipple. Presumably a normal child 
will still feel hungry, although I don't know how much air he has 


At the beginning of his interesting communication Dr. Monnier 
said that he did not consider that growth could be determined by 
generalization because generalization is a primitive stage which 
comes before differentiation. I think there is some ambiguity, not in 
what Dr. Monnier said, but in the discussions in general, particularly 
in the explanations given by Pavlov's followers of the formation of 
language by conditioned reflex. Actually this term generalization can 
be taken in two ways, which appear to me to be very different and 
which must be carefully distinguished if we do not want to become 
involved in confusion associated with the definitions of psycho- 
biological terms. 

The word 'generalization' can be first taken in the sense used by the 
reflexologists as being a level of initial non-differentiation. Generah- 
zation, in the sense of a reflex, is a level where reactions are not yet 
differentiated or specific. I would call this type of generalization 
'automatic generalization'. 

But there is a second kind of generalization which plays an impor- 
tant role in growth and in the development of intellectual functions, 
which I would call 'intentional generalization'. Contrary to automatic 
generalization it depends very much on differentiation; the two are 
in fact interdependent. 

I will give an example. One of my children at about four and a half 
months, an age when he was beginning to take hold of objects within 
his visual field, in the early stages of co-ordination between vision 
and prehension, started to take hold of a cord hanging from the 
canopy of his cot. When he took hold of the cord it shook all the 
doUs which we had hung from the edge of the canopy and this 
immediately gave rise to a circular reaction, a plan of action which he 
kept on repeating. In this case the plan gave rise to generalizations 
but these were accompanied by very clear differentiation. When a new 
doll was hung from the canopy you could see the child looking for the 
cord and puUing it while looking at the doU, but if a doll was held in 
front of him he did not look for the cord, he had other reactions. 
Later on I tried the experiment of hanging an object not from the 
canopy but from a long rod about two yards away from the child. 
I made it swing. The child looked at this with great interest and 
smiled, but the moment the movement stopped he looked for the 
cord because of the hanging object. Here we have a generalization 


which is a sign of development, a generahzation based on differentia- 
tion and discrimination, an active and intentional generalization 
which should not be confused with automatic generalization. 

If it is a question of automatic generalization, then I am in full 
agreement with Dr. Monnier, but not if it is a matter of intentional 

I am insisting on this distinction because in Pavlovian discussion 
on the formation of language based on conditioned reflexes the great 
difficulty is to explain the generahzation inherent in the use of words, 
nouns and concepts. Very often there is a change from the first 
meaning to the second and from the second to the first. I repeat that 
according to the second meaning generalization is a very clear sign 
of development of everything connected with cognitive functions. 


I wonder what one would possibly derive from considering the 
newborn's ipsilateral and contralateral reaction movements of the 
head in relation to the genesis of object relations. We have apparently 
a changeover from what we might call the ipsilateral reaction 
to the contralateral reaction, and at a certain time we even had, 
according to this film, some sort of conflict situation between 
ipsilateral and contralateral tendencies. Now I hesitate to give a 
psychological interpretation to the ipsilateral reaction, but I think we 
cannot go far wrong if we consider the contralateral reaction as a 
flight reaction from something unpleasant. If one considers the 
matter from this point of view, it would seem to me that it bears out 
an idea as to the genesis of the concept of object. It would seem 
reasonable that in the beginning the newborn has no concept of 
object at all; he considers the object part of himself. It seems to me 
that in all likehhood the genesis of the concept of object is related to 
the first experience of objects which are not dominated by him or 
belonging to him, that is to say, to introduce a term of Melanie 
Klein's, bad objects : objects which are opposed to the newborn and 
which, therefore, for the first time put him up against the necessity of 
considering that there are certain things in this world which are 
hostile and against which the better choice is a reaction of flight. I felt 
that Dr. Monnier's film showed very clearly not only how the ipsi- 
lateral embracing of the object expressed the all-belonging quahty, but 
also how the contralateral reaction, the first awakening of the concept 
of the hostile object, arose, and how the two entered into conflict, 
which in all likehhood is one of the first conflicts which the newborn 
experiences. This I feel might be an interesting aspect of the problem 
of neurotic conflict later on in life which can be derived from the 


purely physiological considerations and builds a bridge towards 
psychological and psychoanalytical concepts of this phase of 


I do not know if we are entitled to bring primitive ipsiversive 
patterns in relation with the psychological concept of all-belonging 
behaviour and the contraversive pattern with the concept of hostile 
rejecting behaviour. Of course, in our anencephaUc newborns, the 
finger is not recognized as an object. It is just a source of stimulation. 
We must ask Professor Piaget at what time the notion of an object 


That depends on the criterion you use for the idea of object. If 
you take as criterion the search for an object which has disappeared 
from the perceptive field ; for example, if you put a watch in front of 
a baby and just at the moment when he is about to take hold of it 
you cover it with a cloth, the reaction remains negative until about 
six or seven months. It is only between eight and ten months that we 
find the search for an object which has disappeared behind a screen. 
Before this behaviour is organized, very interesting intermediate 
stages can be observed where the object is not yet localized. Thus, if 
you put a baby between two pillows, one on the left and one on the 
right, and you place a watch under the right pillow (you do this at 
the time when the baby is beginning to search behind a screen) the 
baby lifts up the right pillow and takes the watch. Then in front of his 
eyes you put the watch under the left pillow. I have seen infants of 
eight or nine months at the moment the watch disappears looking 
again on the right where the action was successful the first time. 
Therefore there is not yet permanence or localization of the object. 
Consequently if you take as criterion an object which is localizable 
outside the perceptive field only towards the end of the first year do 
you get search for an object that has disappeared. 


In the light of present experience it seems impossible to fix on an 
exact time, say four and a half months, for the sensori-motor 
co-ordination necessary for prehension of the object. There is always 
a certain learning in the act of taking hold of an object even when the 
object is within the child's visual field. 



This happened at diflferent ages in my three children. In the first 
it was at four and half months, in the second at six months and a few 
days, in the third at three months and a few days. With the third I 
had carried out a series of experiments on imitating hand movements 
which seem to me to have played a part in this early reaction. 


I want to pose the question of the Spitz smihng reaction. It seems 
to me that is the first evidence of the perception of objects. What is 
Dr. Bowlby's view about that? 


I think we have to distinguish between responding to a perceptual 
stimulus as an Innate Releasing Mechanism does, and the abihty to 
conceive of an object persisting even when it isn't being perceived and 
responded to. I should have thought these were quite different things, 
and that Spitz was talking about the I.R.M. response principally, 
and that his work indicates that it isn't until after six months that an 
infant develops the notion of a persisting object. 


I think the smiling reaction is a very complex phenomenon. The 
origins of imitation are still rather mysterious. Eight years ago I 
observed a fact which I consider almost inexplicable. I noticed in my 
son, who was then twenty days old, the imitation of putting out the 
tongue. At first I thought it was a mistake, that it was I who had 
imitated the child and not the child who had imitated me. I tried this 
experiment again several times. I took a film of it and later I was able 
to examine six infants in whom this imitation also appeared between 
the twentieth and thirtieth day, then disappeared, and reappeared at 
the age of about three months. I have formulated some hypotheses on 
this subject but they seem to me not very sound. I tell you about this 
fact because I should like to know if any of you have noted such early 
signs of imitation. 


Is the imitation which appears at three months lasting ? 


It is lasting on condition, of course, that it is cultivated. You hold 
the child, he is close to your face, he concentrates on your mouth, 


He starts putting out his tongue when you do ; he smiles when you 
smile ; he starts again when you start again. The imitation of pro- 
trusion of the tongue then again becomes contemporaneous with the 
smiling response. But there is that very early imitation at the age of 
twenty days which is strange and which perhaps challenges our over- 
intellectual theories on imitation. 


The athetotic movements which are seen in prematures are very 
reminiscent of stretching movements with young. Stretching, like 
many other instinctive activities, has a very definite gradation of 
intensities and, at the highest culmination of intensity, it is always 
accompanied by yawning. That this very same thing occurred during 
the athetotic movements of your anencephalic in the film seems very 
important to me. Furthermore, your friend Professor Gamper, in the 
titles of the film, always uses a word which quite simply means 
stretching, so evidently he himself had no doubt about the funda- 
mental identity of stretching and athetotic movements. In the adult, 
matters seem to be somewhat different. I have seen lots of patients 
with lesions of the pallidum and consequent athetotic movements, 
but the impression that there was a similarity between these move- 
ments and stretching never occurred to me. But it did instantly when 
I first saw the athetosis of a premature. And this impression was very 
strongly accentuated by your anencephalic yawning while doing 
athetotic movements. The beginning of it was definitely athetotic and 
the end was in yawning. 

There is one other point on which I wanted to utter an opinion. I 
agree with Dr. Monnier's interpretation that two mechanisms were 
in conflict, that this creature actually did the 'oral prehension' 
reaction and at the same time the flight reaction too. I think that this 
interpretation is quite correct because conflicts are not in any way 
dependent upon the presence of higher nervous functions. You get 
true conflict on the level of the medullary preparation. It is the conflict 
between two postural automatisms. Instinctive actions or instinctive 
activities of any kind primarily overlap. It is not the general rule that 
they are mutuaUy inhibitive. They are mutually inhibitive only in 
special cases where overlapping would be harrnful. Flight reaction 
must not be overlapped by any eating reaction. If an animal running 
away did a mixture between running away and eating it would be very 
bad for him. The overlapping of sexual and aggressive reactions in 
fishes, which we are investigating just now, is so common that you find 
the greatest difficulty in ascertaining which movements are motivated 
by the sexual instinct and which are motivated by the aggressive 


instincts. Judging by all we know, the factors which keep the several 
instinctive activities apart and prevent them from going off simul- 
taneously, in a 'cacophony of movement', as Tinbergen expresses 
himself, or, more simply, in a fit, lie in the afferent control. There is 
no doubt that the greater part of motor mechanisms are localized 
farther to the caudal end of the brain than are the controlling afferent 
mechanisms. In the anencephalus, devoid of all but the most caudal 
parts of the brain, an uncontrollable overlap of motor activities is 
therefore only to be expected. 


I think that we must distinguish the athetosis of the trunk from the 
athetosis of the extremities. The latter appears chiefly when the 
inhibitory action of the striatum, and more precisely the action of the 
caudate nucleus, is abolished, as a consequence of extensive lesions. 
The lower centres of the brain stem deprived of the striatal control are 
then released and show an increased activity, in the form of athetoid 
movements of the face, fingers, and toes. Athetosis of the trunk 
develops in a similar way in the rhombencephalic anencephalus, when 
integrated structures of the tegmentum and of the reticular formations 
of the pons and medulla are released from the diencephaUc control. 
The athetoid stretching of the spine and neck, with yawning and 
worm-Hke contortions of the body, are mechanisms which must be 
integrated in the tegmentum of the midbrain and pons, as well as in 
the reticular formation of the medulla. 

As for the question of conflict between ipsiversive and contra- 
versive patterns, I agree entirely with Dr. Lorenz. Both patterns 
must have at the same time a diff'erent, well-integrated anatomical 
substrate, with different thresholds; the excitability of each sub- 
strate may vary according to the degree of maturation and to the 
various endogenous conditions. A conflict situation occurs because 
the afferent controls are hmited to spino-bulbar systems, in the 
rhombencephalic anencephalus, whereas in normal newborns 
cerebellar diencephalic and cortical controls, with their afferents, are 
still acting. 



Criteria of the Stages of 
Mental Development 


I find myself in a most unenviable position. To begin with, I have been 
asked to expound Piaget's conception in front of Piaget himself. The 
conception of mental development as it appears in the works of 
M. Piaget is somewhat disconcerting, not because of the facts but 
because of the terminology. M. Piaget, who is a zoologist by training, 
an epistemologist by vocation and a logician by method, employs a 
terminology as yet not much used in psychology (Piaget, 1951a, 
1951b, 1952). He expresses himself mainly in terms of structures, 
which by definition are systems of mental operations obeying 
definite laws of composition such as, for example, the mathematical 
laws of group and lattice. According to a number of cyberneticists 
structures are as much physiological as mental. It seems to me neces- 
sary to keep in mind this triple orientation — biological, epistemo- 
logical and logico-mathematical — which is continually reflected in 
Piaget's vocabulary, in order to find one's way easily among the 
Geneva studies. But once these characteristics are appreciated the 
data and laws deriving from them become clear and are easily 

The general subject of this meeting is a determination of the criteria 
of the stages of development. How can we define a stage of develop- 
ment from the psychological point of view? The schools of Freud, 
Wallon, and Piaget have adopted different but complementary points 
of view. I should like to expound briefly that of M. Piaget. 

The stages are defined by two main criteria : 

(a) the process of formation or genesis ; 

{b) the complete form or final equihbrium. 

The equilibrium of a stage while marking the completion of one 
period marks at the same time the beginning of a new period of 


M. Piaget has been able to demonstrate three types of structures: 

(1) sensori-motor 'group' structures; 

(2) concrete operation 'groupemenf structures; 

(3) combined 'group' and 'lattice' formal structures. 

I will define each of these structures while describing them. 

1. The structure consisting of a group of sensori-motor operations 
appears in the period of infancy and prepares for the stage of child- 
hood. It is achieved at about one and a half years and is characterized 
by a system or group of reversible actions. The term 'reversible' is 
taken not in its medical but in its mathematical sense. Thus at one 
and a half years the baby is usually capable of making detours and of 
retracing his steps : in other words, of carrying out what Poincare has 
called a 'group of displacements'. However, these actions are carried 
out by successive movements and not yet with the help of simul- 
taneous representation. 

2. The structure of concrete groupement begins in early childhood 
and reaches its equihbrium during later childhood. Actually it comes 
to full achievement between seven and eleven years. The concrete 
groupements which are carried out mentally permit the simultaneous, 
and not merely successive, evocation of a displacement or transfor- 
mation and its inverse. Thus, for example, when a child transforms a 
ball of plasticine into a sausage or a cake (Fig. 7*) he can from seven 
years onwards mentally cancel this transformation and thus arrive 
at the conservation of matter. At about ten years he shows himself 
capable of carrying out the same reversible reasoning in connexion 
with conservation of weight and at about eleven years with conserva- 
tion of volume. In each of these reasonings an actual transformation 
is cancelled by an inverse mental operation, thus leading to conserva- 
tion. However, the child, unlike the adolescent, can only carry out one 
operative groupement after the other, not two simultaneously. 

3. The structure of combined groups and lattices of formal thought 
marks the peak of adolescence. This structure develops between 
eleven and fourteen years and reaches its equihbrium at about fifteen 
years. The groups of formal operations integrate the partial groupe- 
ments in a structured whole. The adolescent carries out a group of 
formal operations of the lattice type when he makes combinatorial 
analyses. At about fifteen years adolescents can make up a mixture 
of a number of chemical solutions not merely by chance but through 
combinations, associating each of the elements with all the others 
of the system. This reveals a new structure. Similarly, the adolescent 
can carry out a number of formal operations with a reciprocal 

* I wish to thank Mr. Vinh Bang for the accompanying illustrations. 


FIG. 7 


and a negative corresponding to each, thus showing a group struc- 
ture. The concept of proportion which the adolescent appHes in the 
field of geometry and physics depends also on the group. The 
structure of formal operations thus shows an unhmited degree of 
reversibihty and mobility. 


Having defined the structures we can now describe the stages of 
development as processes of formation leading to structures of 
equihbrium. I will limit myself to giving a few brief indications and 

The first stage stretches from zero to one and a half years approxi- 
mately. This first period of life can be characterized by the genesis 
of the sensori-motor stage of intelligence and is achieved with the 
formation of the sensori-motor group. This consists of a combina- 
tion of reversible actions such as displacements in space. Six sub- 
stages mark the gradual progress of development during this first 
period of life, with a gradual extension and increasing mobility of the 
'schemata' of behaviour.* 

* Piaget calls schema a piece of behaviour which can be repeated and co- 
ordinated with others. 


(1) 0-1 month: Reflex exercises. 

(2) 1-4| months : Primary circular reactions (formation of motor 
habits and perceptions). 

(3) 4|-9 months : Secondary circular reactions (formation of in- 
tentional acts and prehension). 

(4) 9-11/12 months: Co-ordination of schemata (ends and means) 
and constancy of the object. 

(5) 11/12-18 months: Invention of new means (sensori-motor 

(6) 18 months: Internalization of the sensori-motor schemata and 
achievement of the group of displacements (detours). 

The second stage is characterized by a period of formation and a 
period of equilibration. The period of formation is from one and a 
half to seven years and the period of equilibration from seven to 
eleven years.* 

The formation period is characterized by the genesis of repre- 
sentative intelligence. Within this long period of formation can be 
distinguished two phases without clear demarcation. 

The first is determined by the formation of symbohc thought 
leading to representation. Actually the change from sensori-motor 
action in the infant to mental representation in the child is due to the 
symbolic function which differentiates the significant from the 
significate. Everyone knows of the child's first attempt to represent 
events by symbolic play, drawing, and language. 

The second phase is determined by the formation of concrete 
operations. Mental actions (internalized actions accompanied by 
representation) are irreversible before being grouped in reversible 
systems. Up to the age of six years the whole intellectual behaviour 
of the child is still determined by the irreversibility of mental actions. 

* Since every equilibration phase is at the same time (in respect of the later 
phase) a preparatory phase, the phase seven to eleven years can be considered 
equally well as the phase of equilibration of the structures prepared between two 
and seven years and as the preparatory phase for the structures which are com- 
pleted between eleven and fifteen years. Therefore we should speak of the stage 
from two to eleven years and another from seven to fifteen years, because struc- 
tures follow each other with no definite break, since each new structure integrates 
the preceding ones and is prepared by them. The break at seven or eleven years is 
simply a question of convention or convenience. Here we use the following con- 
vention: stage I from zero to one and a half years; stage II from one and a half 
to eleven years with formation or genesis from one and a half to seven years and 
equiUbration from seven to eleven years ; stage III from eleven to fifteen years 
with final equilibrium reached at about fourteen to fifteen years. The levels of 
equilibration for concrete operations (particularly at seven years and eight to ten 
years) are numerous precisely because these operations are not yet formal, that 
is to say they are not yet entirely detached from their contents but constitute a 
progressive structuration of these various contents. 


FIG. 8 


I will quote as an example the behaviour of two boys, Vincent and 
Marco, who at regular six-monthly intervals were willing to undergo 
a psychological combined with an electro-encephalographic examina- 
tion. We wished to complement our cross-sectional studies undertaken 
upon a large number of subjects by some longitudinal studies. The 
psychological behaviour of Marco at the age of five years six months 
and of Vincent at the age of six years is in fact marked by irreversi- 
bility in their reasoning. Here is a sample of the experiments which the 
two boys were submitted to. 

For the first test (Piaget and Szeminska, 1941) a certain number of 
egg-cups and eggs are used (Fig. 8). The two boys had not the 
sHghtest difl&culty in choosing from a basket as many eggs as there 
were egg-cups on the table. By means of an operation of a one-to-one 
and reciprocal correspondence they were able to place an egg each 
time opposite an egg-cup and so on. However, when the experimenter 
destroyed this perceptual correspondence by spacing out the eggs 
and putting the egg-cups close together or vice versa the children 
denied the existence of conservation of number and estimated the 
number of eggs as a function of the space occupied, 


FIG. 9 




^js- '-3b- 

In a second test (Piaget and Szeminsbca, 1941) a liquid has to be 
poured from a tall, narrow glass into a low, wide one or else from a 
big glass into several little ones (Fig. 9) (for children we make beheve 
it is a fruit juice we want to pour). Although they had themselves 
poured the liquid, the two boys beheved that its quantity increased 
or decreased. The direct action of pouring could not be reversed 
mentally. In the same way the two boys were incapable of under- 
standing the reciprocal compensation of dimensions (tall x narrow 
= wide X low). 

FIG. 10 

For the third test (Piaget, Inhelder and Szenqnska, 1948b) two 
rods of the same length are used. One of them is displaced parallel 
to the other (Fig. 10). Once again the two boys, centering their 
attention on the displacement of the rod, through lack of reversibihty, 
thought it became first longer and then shorter than the other. 

In the fourth test (Piaget, Inhelder and Szeminska, 1948b) two 
surfaces are used which represent fields on which two cows are 


FIG. 11 

pasturing (Fig. 1 1). On each of the fields simultaneously we put a first 
house, then a second house and so on up to fourteen ; only on one of 
the fields the houses are put touching each other whereas on the other 
the houses are spread out over the whole field. Now the problem is : 
Are the unoccupied areas of equal size. In the language of the child we 
ask: 'Have the two cows still got the same amount of grass to eat?' 
Here again it is because of lack of reversibility in their mental actions 
that the children are incapable of seeing the equality of the remaining 

For the fifth test (Piaget, Inhelder and Szeminska, 1948) a lake is 
represented with islands of diff"erent sizes on which the child has to 
build houses of the same volume 'with the same space inside' (Fig. 12). 
For the young children there is no question yet of making three- 
dimensional measurements. What interests us is to know whether or 
not they can think of compensating unequal dimensions. The two 
boys invariably constructed all the houses of the same height what- 
ever the area of the base. 

An absence of reversibility goes along with a certain rigidity in the 
systems of reference. This is why during the sixth test (Piaget and 
Inhelder, 1948) Vincent and Marco were not yet able to imagine the 
water level as being horizontal in inclined flasks (Fig. 13). 

In respect of a seventh test (Piaget and Inhelder, 1941), and an 
eighth (Inhelder, unpublished), the two children had difficulty, 
which is characteristic for their age, in arranging objects in series 


FIG. 12 

according to their size, or in classifying them according to two or 
three criteria at once. 

In short, in these few tests, and in others, the two children were 
capable of carrying out mental actions but not yet mental operations, 
operations being by definition reversible mental actions. 

After a slow continuous evolution the change from irreversibility 
to reversibility often occurs abruptly for a particular problem. 
Concrete operations as a whole, however, only very gradually impinge 
upon reality; the age of seven years marks only the beginning of 

The change from irreversibility to the first forms of reversibihty 
occurred quite suddenly in Marco. After an interval of six months, 
that is to say between six years and six-and-a-half years, Marco's 
behaviour when confronted with the same experiments was com- 
pletely altered. By means of a system of reversible operations he was 
able at six and a half years to understand certain invariances which 
he had denied at the age of six years. In the same way he was able 
to carry out operations of arrangement in series and of classification. 
The same change was more gradual but not less striking in Vincent's 


FIG. 13 


The balanced structure consisting of a groupement of concrete 
operations not only marks the conclusion of early childhood but 
serves as a basis for further development. Here can be distinguished 
different substages. At the age of seven years on the average the child 
is able to carry out logico-arithmetical operations (classifications, 
arrangements in series, and one-to-one correspondences) but it is 
a year later that the time-space operations are achieved (Euchdean 
co-ordinates, projective concepts and simultaneity). Thus up to 
about eleven years there develops gradually a system of concrete 
operations which will later serve as a basis for formal operations. 

The third stage is characterized by the formation of formal opera- 
tions which reach their state of equilibrium at about fourteen to 
fifteen years. At eleven years the pre-adolescent is already capable of 
deducing by means of hypotheses and not simply from concrete 
facts. His reasoning frees itself from the concrete. But only at about 
fourteen to fifteen years does this new form of intelligence attain a 
balanced structure governed by laws of groups and lattices. 

In fact two boys, Philippe at the age of fourteen, and Udrea at the 
age of fifteen and a half, showed themselves capable of carrying out 
combinatorial and proportional operations, whereas a year earlier 
confronted with the same experiment they only proceeded by trial 
and error without reaching an exact solution to the problems. In the 
experiment mentioned earHer, developed with the help of Dr. G. 
Noelting (Inhelder, unpublished) on colouring obtained by mixture 
of different chemical solutions, the two boys proceeded through a 
systematic combination of the elements presented. They combined 
the five colourless solutions in different orders : 1 + 2, 1 + 3, 1 + 4, 
1 + 5 ; then 1+2 + 3 and so on, with two, three, four, and five 
elements until they were able not only to obtain the colouring asked 
for but to discover the part played by a neutral and a reversing 

In the projection experiment developed with the help of Mr. Vinh 
Bang (Inhelder, unpublished) (Fig. 14) the two boys managed, at 
fourteen and fifteen years but not eariier, to produce a single shadow 


FIG. 14 

on the screen by means of a series of rings of different sizes placed at 
different distances, discovering, without previous teaching at school, 
that the size of the shadow is proportional to the diameters of the 
rings and inversely proportional to the distances from the source of 

The formal type of reasoning of adolescents is thus disclosed, not 
only through verbal expression, but also by the way they organize an 
experiment and furnish a proof. The age of fourteen to fifteen years 
seems characteristic of this last form of equilibrium which brings 
about the completion of formal operations. 

Criteria of Stages 

In conclusion I would like to specify the criteria of the stages. 

1 . The stages of development are defined by structured wholes and 
not by any isolated pieces of behaviour. 

The concrete groupement structure allows not only the solution of 
particular concrete problems but all the elementary types of classifica- 
tion, arrangement in series, and conservation of number. The 
appearance of a structured whole allows us to generalize from one 


particular piece of behaviour to others of the same type. UnHke the 
tests modelled on the Binet-Simon tests which do not allow of any 
generaUzation since they proceed by summation of successes and 
failures, the appearance of an operational groupement allows us to 
identify a mental structure. 

But there is more than that: structured wholes go beyond the 
operations actually carried out and are the base for a whole system of 
possible operations. We have seen that when confronted with the 
problem of combining chemical solutions the adolescent at a 
fifteen-year level proceeds to use a combinatorial method without any 
previous teaching. Thus not only does he recall the operations already 
carried out but he can construct a system of possible operations. 

2. The passage from an inferior stage to a superior stage is equivalent 
to an integration: the inferior becomes part of a superior. It is easy to 
show that concrete operations serve as a base for the formal opera- 
tions of which they are part. The combinatorial method, for example, 
is based on changes of order which are possible during childhood and 
later develop into combinatorial operations. Proportions themselves 
are operations applied to operations, or operations to the power 

3. The order of succession of stages is constant but the age at which 
the structures appear is relative to the environment, which can either 
provoke or impede their appearance. The genetic development 
seems to follow a general law of the same type as the laws of organic 
growth. However, may I emphasize this: the age of reahzation cannot 
be fixed absolutely; it is always relative to the environment. The 
influence of the environment can act in many ways — at one time 
through the content to be constructed, at another by the possibihties 
of learning, or again by the social interchange itself. 

The content to be structured : a group of objects may be more 
or less easy to classify according to their particular perceptual 

Learning : it has been found that certain spatial representations are 
made easier by sensori-motor explorations. 

Social interchange: certain comparative studies have shown that 
in an environment of free interchange and discussion magical repre- 
sentations decline rapidly in favour of rational representations, 
whereas they persist much longer in an authoritative environment. 

These observations as a whole show the age margin which must be 
allowed for in our stages. Even if the intellectual development 
follows a constant order its manifestations are subject to fluctuation. 

In summary we could say that the criteria of stages as shown by 
M. Piaget are based on structured wholes which follow one another 
in a constant order according to a law of integration. 


The genetic conception of M. Piaget opens a number of new 
perspectives : 

(1) The operational pattern of psychological structures may per- 
haps facilitate correlation with the neurological (cybernetic) patterns. 

(2) Since the development of cognitive functions cannot be dis- 
sociated from that of affective functions, it will perhaps be possible to 
demonstrate their parallehsm. M. Piaget has already shown the 
relation between the intellectual operation and social co-operation, 
as well as the interdependence of the pre-logic of childhood and 
moral reahsm. 

(3) The establishment of a scale of development based on balanced 
structures will enable us to identify the level of operations in a 
child and not only individual successes and failures. 

(4) The study of structured wholes is, however, insufficient if it 
is not complemented by research in differential psychology (sex, race, 
social environment). 


Thank you very much. I watched Professor Piaget's face very 
carefully and he seemed calm and peaceful and at times even 
dehghted. Mile Inhelder's presentation is now open to discussion. 


I have been, in England and in my particular miUeu, one of the 
most enthusiastic proponents of M. Piaget. I have tried to convey the 
ideas which have been developed here to my colleagues in physiology, 
with varying success, and I should like to put to you directly a ques- 
tion which is always put to me when I am trying to convince my 
colleagues that this type of behavioural analysis has vahdity in 
connexion with physiological problems. It is this : could you tell us 
very roughly how many children you studied, from what groups 
they were drawn, and whether you have subjected your quantitative 
results to any of the standard statistical analyses ? 


I am not able to quote from memory the exact number of subjects 
examined in each of our studies. For the studies deahng with 
experimental reasoning in children and adolescents we examined 
individually more than 1,500 subjects from five to sixteen years. In 
some studies the examination of 100 children was enough to give us 
interesting indications, whereas for others 200 to 500 were necessary. 

Moreover we are now taking up again with our students those 
studies which gave us the most significant results for the diagnosis 


of mental development. We are working on a large scale and trying 
to study for particular age-groups the relations between our various 


What I would like to have is, for example, a set of distribution 
curves showing the range of the variation of these various behaviour 
standards with age, comparable with the curves for reading abiUty or 
arithmetical ability or the other factors which more conventional 
psychologists are accustomed to plot. 


One of our students is now preparing a thesis for a doctorate on 
this subject and he is studying the distribution of the results as a 
function of age. 


All this work reminds me very much of the sequence of ossification 
occurring in children. Obviously, this is a large jump in analogy. 
Nevertheless, something similar does happen. The sequences of 
ossification are held to even if the child's development is slowed up. 
If I understand Mile Inhelder and Professor Piaget correctly, one of 
the most cogent arguments for the existence of their developmental 
stages is that the sequence of them remains the same even if as a 
whole they are retarded or advanced. This is exactly the same with 
ossification centres, and this seems to me a powerful argument in 
favour of the existence of the mental stages, and of their neurological 

I want also to ask a question. Some children are advanced physically 
during the period of growth, and we have several ways of measuring 
the degree of advancement, of which two, I think, are of chief interest 
in this context. One is the state of the ossification centres — the 'bone 
age', and the other is the state of development of the teeth. At the 
present time my colleagues and I are doing a study in which we hope 
to relate these two things — to discover whether a child that is dentally 
advanced is also skeletally advanced. In general I am fairly sure that 
the two measures are not related to the same thing, though there 
may be a small correlation between them. The head grows rather 
differently from the rest of the skeleton, and the teeth are part of the 
head. Now does advancement or retardation in the sequence of 
mental tests relate to advancement or retardation in the teeth, or 
in the skeleton, or in neither ? 


In the present state of our research I am unable to reply to that. 


It seems to me that the main interest of the Geneva work lies in 
the establishment of the sequence and the explanation of the passage 
from one stage to another. This is a contribution of considerable 
importance to psychology. However, one problem arises, that of the 
curve of growth, and of the significance of the more or less rapid 
passages from one stage to another. Mile Inhelder moreover stated 
the problem very clearly in her work on mental deficiency and she 
has already formulated several hypotheses. The problem I should 
like to underiine is one of method. You said that the age at which the 
structures are realized is very variable and you give as the main reason 
the influence of environment. This is obviously one of the causes. I 
wonder, however, if there are not others. It has been noted for a long 
time that tests of mosaic or multiple-variety type enable us to deter- 
mine the mental level with greater precision. One wonders whether 
Piaget tests should not be used in conjunction with other tests deaUng 
only with an isolated aspect of behaviour. In all these tests one notes 
a surprising unrepeatability of results which arises, no doubt, from 
the educational conditions, diversity of environment and perhaps also 
from the conditions of the experiment. There are extraordinary 
variations related to the test situation, the experimental situation, and, 
in a general way, affective conditions. There is no doubt that this 
type of test is the most subject to affective fluctuations linked with 
experience. I think, therefore, that the difficulty of establishing a 
precise age for your stages comes not only from the fact that there is 
a wide dispersion but also in certain cases that an isolated activity 
does not enable us to obtain the same results as when a mosaic-type 
test is used. I should Hke to know what Mile Inhelder and Professor 
Piaget think on this last point. 


I agree with M. Zazzo that there is a certain dispersion and even a 
certain lability in intellectual behaviour. Nevertheless, I am always 
struck by the fact that among developing children the dispersion 
within a stage is relatively small compared with the wide differences 
between behaviour in one stage or another, between one mental 
structure and another. 

I am not so pessimistic as you as regards the unreliability of the 
results due to fluctuations in affectivity. It is above all important to 

encourage each child to do his very best by creating an atmosphere 
favourable to the examination. 


It seems tliat with these two types of test, the mosaic test and the 
Piaget test, we have this alternative: with the first type, we may 
determine a very precise age, within three or four months, but without 
knowing anything that lies behind it from the point of view of intel- 
lectual mechanism; with the second type of test, we may understand 
the intellectual mechanisms very well but not be able to fix an 
intellectual level. In the present state of affairs it does not seem 
possible by means of your tests to establish any kind of prognosis of 


I am entirely in agreement with you. In the present state of our 
research we are not able to say: 'such and such a child is exactly at 
the level of six years nine months'. However, I wonder if the com- 
pensations operating in mosaic-type tests are any more than com- 
pensations of a statistical order and whether they reveal the essential 
characteristics of an age level ? That is another question. 


I should like to make a remark to the two previous speakers. The 
object of these studies, initially, was not to estabhsh a scale of 
development and to obtain precise determinations of age as regards 
stages. It was a question of trying to understand the intellectual 
mechanism used in the solution of problems and of determining the 
mechanism of reasoning. For that we used a method which is not 
standardized, a clinical method, a method of free conversation with 
the child. We encouraged each child, as far as possible, in a way 
which was not comparable to that used with the preceding child. 
That is why, personally, I am always very suspicious of statistics on 
our results. Not that I dislike statistics; I worked on biometrics 
enthusiastically when I was a zoologist, but to make statistical tables 
on children when each was questioned differently appears to me very 
much open to criticism as regards the results of the dispersion. It 
would be very easy to make all this into a test-scale, but it would not 
have the value of a standardized piece of work like that which Mile 
Inhelder' s co-workers are undertaking now, for example. In reply to 
M. Zazzo, I think that by taking the operatorial mechanism for a 
particular level one attains something more general than the 


mechanism of compensations in a mosaic test. I think that Mile 
Inhelder has somewhat exaggerated the differences of age, the varia- 
tions with environment, in order to be the 'Devil's advocate'. In 
actual fact, I have personally been quite surprised by the first results 
of Mile Inhelder's students who are now carrying out a standardiza- 
tion in the form of tests. The regularity is greater than I should have 
thought judging by my clinical conversations with the children. I do 
not despair of obtaining a scale which will perhaps not be exact 
within three months but which, from the very fact that it will give a 
structured whole, will reveal more than would a system of com- 
pensations such as the mosaic-tests. 


I should like to know if you have attempted to check the vaUdity 
of your deductions by the 'educational artefact'. I will explain what 
I mean. A child from a known environment who has received the 
instruction normal to this environment is submitted to a course of 
training which might faciUtate for him the acquisition of these stages. 
Have you tried to put yourself in conditions which are different from 
those of the children who are usually around you ? In this connexion 
perhaps Dr. Mead could tell us if among children of a so-called 
primitive population, that is at least not having had a western 
education, one would find the same stages and the same ages defined 
by these stages. Taking the case of Negroes, for example, should we 
not find that the acquisition of these stages in American Negroes 
receiving the same education as the white people was normal, 
whereas on the contrary there would be a marked retardation among 
certain African peoples who have a very different educational 


I have not been able to experiment by varying systematically 
the educational factor or going so far as to carry out the experi- 
ment in a different cultural environment. My personal experience, 
apart from work on normal children, is confined solely to children 
whose schooling has been irregular, children who were refugees 
in Switzerland during the war. I noted that certain of these children 
at first contact gave a response which seemed to be at a level inferior 
to that which one could expect from them. However, after a 
quarter of an hour of experimenting and conversation these same 
children reached a higher level. In a certain sense they had caught up 
on their pedagogic retardation whereas the true mentally defective 
are never able to do this. Therefore it does not seem to me impossible 


that in our cultural environment the method of clinical interrogation 
by means of concrete experiments should enable us to disclose the 
potential level of an individual. But I base this on only a few^ observa- 
tions and cannot draw any generalization from it. 


I would like to give a second reply to Dr. Remond. Let us take the 
last of the stages mentioned, the attainment of the formal level. There 
we have a whole structure which is characterized at first by the 
appearance of the logic of propositions, and if we study the lattices 
intervening in these propositional operations we find a correlation 
with proportions and combinatorial operations. From the point of 
view of the educational artefact it is surprising to note that propor- 
tions are taught, but combinatorial operations, at least in Geneva, 
are never taught in school. However, children invent these combina- 
tions by themselves. They find out not the formula, of course, but a 
complete method giving all the combinations for a certain number of 
variables. Here there are two operations which derive from the same 
structured whole, one of which is taught in school and the other not. 
They are, however, contemporaneous. It is striking that the school 
should have to wait till the age of fifteen before teaching proportions. 
I am certain that if teachers had been able to appreciate the concept of 
proportion in a more concrete manner they would have taught it 
already at eight years. If teachers have delayed this teaching until 
twelve years they have done so with good reason. 


Before I can answer Dr. Remond's question, I should like to ask 
Mile Inhelder if I understood two or three of the points. Do I 
understand that you were working here only with sequences ? Surely 
you have a lower limit below which you do not expect a given stage 
to appear? I would like to know how much possible spread you 
think occurs in development; do you conceive of some children 
reaching the six-year-old expectation at two, or at three or at five? 
Do you think of others not reaching it until ten ? 

The other question is the relationship to what we usually call 
inteUigence. Do you expect a child with an IQ of 70 to show these 
stages in a rudimentary form, or do you assume that it will not reach 
stage 2 or stage 3 at all ? 


As to the first question, I can only reply in a superficial way, 
letting you see later, perhaps, the necessary documents. In one of the 


experiments on conservation solved by 75 per cent of the children 
at the age of six and a half, I did not find a single success until 
exactly five years and no failure after seven and a half. I was dealing, 
of course, with 'normal' children examined in day nurseries and in 
infant and primary schools of the town of Geneva. For other 
experiments the dispersion may be slightly greater or slightly less. 

As to the second question, I have had the opportunity of examining 
retarded children and mentally defective children with IQ's of 70 and 
less. Among these children the true mental defectives, after a period 
of slow development, were able, at the age of physiological maturity, 
to pass the threshold of concrete operation structures (seven to ten 
years) but were never able to attain even the lower threshold of 
formal structures (eleven to twelve years). The imbeciles never 
reached the threshold of concrete operations. However, the IQ of 70 
obtained by means of mosaic-type tests does not appear to me to be 
an absolute norm. I have found a few rare children who, despite an 
IQ of 70, were able to catch up on their retardation and to give in our 
tests results of a higher level. The IQ of 70 can in certain cases mask 
either a normal process unfolding slowly or a pathological process 
tending towards an early halt. 


I am going to have to answer Dr. Remond first on theoretical 
grounds rather than from observations. Stated as Mile Inhelder stated 
it, this framework of stages seems to me to be at such a level of 
abstraction that it will probably be applicable to every people we 
know anything about. The sort of cultural variations that one would 
expect to find would be of this order. The Arapesh, a tribe in New 
Guinea, are people who only count to twenty-four, and after that 
say 'a lot' — they say one, two, two and one, one dog, one dog and 
one, one dog and two, one dog and two and one, two dogs — after 
six dogs they get bored. Now among such a people only the veiy 
brilliant are likely to use the same thinking as Piaget stage three, and 
it would take a considerable amount of extra education, and we 
don't know how much, to elicit from the average the same amount of 
facility in stage three that one would find here in this country. On 
the other hand, in another people not a hundred miles away in New 
Guinea and of the same general racial type who can count to 
thousands in their heads, can do the most complicated mathematical 
arrangements, and are exceedingly interested in adjustment to 
reality, I would expect many more individuals of average intelligence 
able to function as in stage three. Now, when I tried to apply Pro- 
fessor Piaget's formulation of twenty-five years ago, it did not work 


because the formulation was far more concretely expressed than this 
formulation now. And when I tried to take magical thinking, the 
kind of animistic thought imputed to young children, and test it out 
among my second New Guinea tribe, then I found no correspondence. 
They were a tribe that had emphasized relationship to reahty and 
factual reporting so intensively that the other type of thinking did 
not appear in children. 

I think we need to distinguish between the cultural evocation of the 
Hnes of thought called primary process and secondary process think- 
ing, and the way in which the sequence of development of these types 
of thinking is correlated with growth. I expect to find such an 
investigation cross-culturally useful, though at times it may be very 
difficult for a European to recognize the forms of thinking found in 
these different cultures. 

The thing I don't quite understand, though, is the relationship of 
this formulation to the formulation of the child's relationship to 
reality, and the attribution of magical and animistic thinking to 
young children. That is, how do the two aspects of culture integrate 
or criss-cross ? 


I can only make a statement concerning the environment in which 
I have worked. In fact we observe an inter-penetration of the pre- 
logical structures on the one hand with the animistic and realistic 
forms of childhood beliefs on the other hand. The appearance of the 
first structures of logical thought coincides with that of rational 
causality in the child. In our environment it seems that there is 
synchronism between these two processes. 


I think that in order to make comparisons between very different 
social environments, as Dr. Mead has done, it is necessary to find a 
system of tests as far as possible independent of language. All the 
tests which I used formerly had the drawback of being essentially 
connected with language. This is what I would call my pre-operational 
period. But if you take our tests on space, here is a field where one 
can find all the operations which can be presented relatively indepen- 
dently of language by a system of drawings, by comparison between 
a given concrete situation and drawings among which the subject 
can choose the true and the false. One should find something in 
common between different civihzations or cultures in the spatial 
field. These spatial operations are not, however, separate operations : 
one can find a whole series of groupements and other operations 


applied to space. As long as no systematic comparison has been made 
between different cultural environments with these spatial tests one 
has great difficulty in separating the part played by language, with all 
its cultural significance, and the part played by operations. It seems 
to me this still remains to be done. 


We had an interesting experience with the application of the 
Piaget methods in the clinic. It is an anecdote rather than a report 
of the experiment, but it is of interest in relation to whether or not 
certain stages of development can be concealed or evoked by treat- 
ment or circumstance. We had some adult psychiatric patients who 
displayed extreme immaturity, with retardation in their E.E.G.'s and 
other physiological peculiarities, and we attempted to relate this 
finding with the stage of thinking. We did a series of tests which we 
based entirely on Piaget principles, and found, in fact, that patients 
who were in a severely disturbed psychopathic state had 'retarded' 
E.E.G.'s, and were very primitive in their behaviour, were not in 
stage three as they should have been, being adults, but in stage two 
of their development. 


Had they never been normal? 


Yes, they had been normal people. They were aged between 
twenty-five and thirty and they had been perfectly capable up to a cer- 
tain point, but they had regressed apparently back to a state where 
they could not do tests equivalent to simultaneous equations; they 
couldn't make reciprocal reversible relationships even in very simple 
tests. Some of them were treated and some got better and we could 
observe, in fact, a correspondence between the electroencephalo- 
graphic changes and re-maturation and reacquisition of the capacity 
for formal reciprocal thought for the solution of simultaneous 
equations. Now we have just applied exactly the same test to a 
population of delinquent children. The application of Piaget-type 
tests to a delinquent population aroused such an intense emotional 
response in the subjects that the people responsible for the children 
suggested we should abandon the tests forthwith. 


What were these tests ? 



They were very much the same as those Mile Inhelder was des- 
cribing. We put water in glasses, we dropped a stone in the water, and 
so on. A particular study was made of the question of causahty — we 
said. Why do you think the stone sinks ? — and they said, 'Good God, 
if you think I'm such a fool as that . . .' — but they didn't answer. 


As regards the question put by Dr. Mead, I think it is extremely 
difficult to separate the parts played by education and by heredity 
in development because the comparisons that we can make must 
always be carried out, even in primitive societies, in a human 
environment where there is language. Moreover, it appears that 
cultural and genetic factors are not additive. Their relationships are 
certainly much more complex. When reading M. Piaget's works I 
always wondered to what extent evolution and culture derive by a 
kind of maieutic process from hidden psychological aptitudes. In this 
respect, I would recall certain current jokes or certain well-established 
facts. It is said that the child of eighteen months is at the chimpanzee 
age and that at seven or nine he is at the Aristotle age. 

You know of certain cases of wild children brought up by animals, 
notably the famous case of the wolf-child studied by Gesell (1941). 
The report published on the story of little Kamala is very interesting. 
At the time she entered the orphanage Kamala, who seemed to be 
about six or seven years, showed no human behaviour. She was 
re-educated under fairly strict control and at fourteen years, when she 
died from a uraemic crisis, she had reached the stage of language. 
This is a fascinating problem. It is obvious that when she was dis- 
covered in the jungle Kamala had not reached Piaget's second stage 
and yet other examinations showed that neurologically she was 


I am sorry but I must lodge a passionate resistance against Amala 
and Kamala. I'll take my oath, and I want to drop dead this minute, 
if these children have really been raised by animals, and if you try to 
get hold of Gesell — as I did — he doesn't want to talk about it. 
Mr. Singh and Mr. Zingg, I am sorry to say, are people to whom the 
German saying applies, 'Wer einmal liigt, dem glaubt man nicht, und 
wenn er auch die Wahrheit spricht'. 

If somebody assures me that a child raised by wolves has green 
luminous eyes, then I don't believe a word he says any more. A friend 


of mine has caught them out in another untruthfulness. In Zingg's 
book (1941) he proudly refers to an Enghsh scientist who, according 
to him, also refused to believe at first, but later humbly apologized. 
My friend. Dr. W. H. Thorpe, F.R.S., sought out that man, whose 
name I have forgotten, to ask him why he apologized and what had 
made him change his opinion. That poor fellow went wild with 
passion; it turned out that he had never apologized at all and did not 
believe a word of the whole story. 

Now let me put before you a few of my arguments why I don't 
believe it. Supposing you have a wolf bitch who has lost her litter — 
if she had lost her fitter there would be some chance of her caring 
for a child. She will grab that child and carry it to her lair, and 
then she will throw herself down and make herself ready to be 
sucked. That is all she does. She has no possibifity of helping the 
child to find her teats. The child must be at an age where it doesn't 
yet grasp, because Amala and Kamala are reported to have eaten 
from the earth without using their hands (which a dog does by 
the way, it does use its hands in gnawing the bone, which Amala 
and Kamala surprisingly didn't, because neither Zingg nor Singh 
knew that dogs did). Then supposing that that child, by some 
incredible accident, happens by rolfing about to find the teats, or that 
the she-wolf, by rolling about also, happens to bring her teats in 
contact with the child and raise that child, the she-wolf would suckle 
that child for two months, and then cease suckfing it and feed it 
on regurgitated carrion. You must remember that she has to start at 
an age when the child still does not grasp or walk. And now I ask the 
paediatricians who are here what child taken by the she-wolf — let's 
be very generous and say at four months of age — suckled two months, 
and then fed by bitch- vomited carrion — what child would stand that ? 


Are all these details on the behaviour of wolves based on Indian 
wolves ? 


Well, those are slightly shorter than the European ones, but other- 
wise they are the same. 


And we have well authenticated details on their nursing and 
feeding ? 



Oh, yes. 



I have a rather long-winded question. It has to be long-winded in 
order to make my point clear. I am somewhat a heretic in regard to 
Gestalt psychology and it is my contention that Gestalt, in its most 
highly developed and most complicated forms, is nothing else but a 
phenomenon of constancy. When I see this paper in white colour, 
this perception is done by a very complicated apparatus working 
absolutely hke a computer, subconsciously, and inaccessible to self- 
observation. This apparatus takes into consideration all other 
objects within the visual field, and assumes, as a working hypothesis 
which may not always be correct, that all these objects are reflecting 
light and colours, with no preference for any particular colour of the 
spectrum. On this assumption, the colour-constancy computer deter- 
mines the relation between the colour of the impinging illumination 
and the colour reflected by the paper. This relation is, within certain 
limits, a constant, and this constant is characteristic of the object. 
This paper reflects, indiscriminately, all the colours of the illumina- 
tion prevailing in this room; in other words, this paper is white. If the 
basic assumption of the colour-constancy computer is wrong, if the 
objects within the visual field do not reflect diff'erent wavelengths 
impartially, but some more than others, the conclusions necessarily 
are erroneous too. If the greater part of the visual field is filled by red 
objects, the computer assumes that it is the illumination which is 
red and concludes that a paper indiscriminately reflecting all colours 
of the spectrum, a white paper, must reflect green light in preference 
to red, in other words, that this paper is green. This miscalculation, 
logical in itself and only based on an erroneous premise, is the cause 
of what is known as the simultaneous contrast of colours. 

Now consider what compHcated operations the computer under- 
lying our perception of form is able to make. They amount to the 
highest operations of stereometry. When I turn my pipe before my 
eyes its image assumes, on my retina, an infinite variety of different 
forms. Yet the form-constancy computer correctly interprets all these 
changes of form in the retinal image as movements of the object, and 
I perceive the form of this object as being perfectly constant. This 
amazing performance is even independent of the perception of depth! 
If I cast the shadow of the pipe on a screen, its rotation is still clearly 
perceptible, only the sense in which it turns has become ambiguous. 
Now let's go one step further and see how much more complicated 


the function of our perception of form and movement can be. Sup- 
pose that my pipe, while I turn it around, should suddenly waggle its 
stem. I should notice it instantly — and wouldn't I be surprised ! Sup- 
pose a duck, swimming before me on a pond, turns around ; its image 
would be foreshortened, yet it would not appear shorter and fatter. 
Now suppose that that duck, while turning, fluflFs out its feathers, so 
that it really would become relatively shorter and fatter. Even in this 
case, which seems confusing even in the telling, the two computers of 
form constancy and movement perception still succeed in keeping 
change of form and movement apart, though they have only the 
retinal image on which to base their computations. Our perception 
has not the slightest difficulty in interpreting the changes in the 
retinal image which are due to movements as movements, and those 
which are due to changes of form as changes of form. Now just hold 
this in mind for a moment and consider how complicated the 
apparatus underlying all these discriminations must be. 

In phylogeny, it often happens that an apparatus, devised for quite 
definite and narrow ends, may prove useful in a quite different and 
much more generalized way of application. The human hand, 
evolved exclusively in the service of climbing, is the most common- 
place example of such a change of function. Something very similar 
has happened to Gestalt perception. Maybe it is quite useful to 
remind English-speaking scientists, that the word Gestalt primarily 
is very nearly synonymous with form. And in phylogeny Gestalt 
perception has quite indubitably evolved out of the function of the 
form-constancy computer. Or, maybe, it is more correct to say it 
evolved out of all the constancy computers collaborating to build up 
our perceptual world. All these computers are objectivating in the 
most hteral sense of the word. That is to say they always tend to 
determine qualities which are constant in and characteristic of the 
object. And that is all that they are reporting to the subject. The form- 
constancy computer does not report the accidental form which the 
retinal image happens to have at the moment, but quite directly the 
real, stereometric form of the object. This disregard for all concrete, 
but accidental, data of stimulation, the coming-down to constants 
essential for the object in question is very characteristic of all the 
constancy computers underlying our perception of things. That is 
what I mean by saying that they are objectivating. There is no doubt 
that all these objectivating functions have evolved phylogenetically 
in the service and under the necessity of recognizing individual 
objects or, to use a term coined by Karl Biihler, in the service of 
'Ding-Konstanz'. But, as the human hand could be turned to new 
functions, the ability inherent in all constancy computers to extract 
essentials enables the organism to achieve something which comes 


very near to the formation of abstract conceptions. It not only 
enables me to recognize my chow Susi independently of accidental 
circumstances by extracting her constant and essential individual 
characters. That process of 'extraction of essentials' can go one step 
further: without any change in the physiological — or, if you like, 
cybernetical — mechanism, my Gestalt perception is also able to 
extract essentials which are not only constant in and characteristic 
of that individual chow bitch, but of all dogs. It is able to disregard 
accidentals which are characteristic only of the individual, and to 
recognize one mutual, unconfusable Gestalt-quaUty of dogginess in 
this chow, as well as in the butcher's mastiff or my old aunt's peke. 
There cannot be any doubt that this direct perception of supra- 
individual Gestalt qualities is functionally closely akin to the forma- 
tion of real abstract concepts. Very probably it always plays a con- 
siderable part in true abstraction and constitutes its indispensable 
basis. Cyberneticists tell me that an analogous possibihty for an 
unexpected change of function has been found in calculating machines, 
too. A machine constructed exclusively for the purpose of calculating 
compound interest surprised its own constructor by being able to do 
differential calculus. The function of this 'abstraction-computer' is, 
after all, not more miraculous than that of the form-constancy 
computer enabling me to see the identical pipe in all those varying 
retinal images. 

Wherever animals achieve something like a generic notion and 
whenever small children correctly apply generic names, they do so 
with the help of the 'abstracting' function of Gestalt perception. This 
is also the reason why little children often form generic conceptions 
with most unexpected contents. I knew a baby whose parents were 
quite desperate because that child 'couldn't tell the difference between 
a dog and a horse', because it said bow-wow to a dog and it said 
bow-wow to a horse and when confronted with a rabbit it said bow- 
wow too, but not when confronted with a goose or any other bird. 
It was clear at once that bow-wow simply meant 'mammal'. My 
friend Verlaine, in Liege, found that monkeys could, in a similar way, 
also achieve 'abstract' conceptions. An old and wise Javanese 
monkey could be trained to recognize birds, mammals, reptiles, and 
insects in coloured pictures and, at last, Verlaine taught him to make 
the difference between live creatures and dead things in general. This 
monkey's performance was so very extraordinary that some critics, 
among them my friend Otto Koehler, suspected that it did not react to 
the pictures that were shown to it at all, but to some signs uncon- 
sciously given by its master, just as the well-known 'talking' dogs and 
horses do. But I do not believe this at all, after having seen that 
monkey at work. With 'talking' dogs, it is always quite evident to 


the knowledgeable observer that the animal does not pay any parti- 
cular attention to the problem with which it is presented, for instance, 
to the characters or numbers that it is supposed to read, but evidently 
concentrates upon its master's face from which the signs guiding its 
performance really emanate. But that old monkey sat down in a 
frenzy of concentration with the picture in its hands, looking at it 
intently, and sometimes actually turning his back on the examiner. 
And after that he gave the correct solution in a statistically significant 
number of cases. 

If you ask my personal opinion I am ready to assert that I quite 
beheve in this monkey's abihty to abstract, by the direct means of 
Gestalt perception, the 'conceptions' of mammals, bird, reptile, and 
so on. And one of my reasons for believing it is that my daughter, 
at the age of five, could do something much more difficult. In my 
own bird collection there were only two representatives of the order 
of Rails, or Rallidae. These two were the Coot and the Moorhen, 
both aquatic birds, adapted to swimming and therefore externally 
similar to ducks. On this statistically very misleading basis of 
induction, my httle daughter's Gestalt computer enabled her to 
recognize any representative of the order of Rallidae. We examined 
her in the Vienna zoo, then containing a very rich collection. The 
Rail birds were kept together with other birds and distributed over a 
great number of different cages. Furthermore, you must know that 
the order in question contains a lot of species which are, in body 
proportions and general appearance, as different from each other as 
could be. There are long-legged forms looking much hke herons, 
duck-like aquatic species, and some dry-land forms that look exactly 
like gallinaceous birds, for instance, like quails or partridges. 
Nevertheless, my daughter picked out the right birds without a single 
error and without any suggestive questioning on my part. On the 
contrary, I tried to mislead her, showing her cages in which there 
were no Rails at all, but some externally similar birds, asking her in a 
voice suggesting that there must be some, but she flatly asserted 
there weren't any! When I asked her how she recognized the Rails 
for what they were, she could only say, 'Well, they are just like a 
moorhen . . .'. 

Now the question to which I am leading up, and which is directed 
particularly at Mile Inhelder, is this: I believe that among your 
intelligence tests for children there are some which investigate the 
child's ability to abstract, to generalize. Do you ever, in such tests, 
find any correlation between a person's power of abstraction and his 
or her particularly ability to perceive complicated Gestalts ? If what I 
suspect is correct, you ought to get in this respect a particularly big 
variabifity in tests of generalizing power, because it is well known 


that the abihty to see Gestahs as wholes is very different in different 
men and different types of men, and if there really is a strong causal 
connexion between Gestalt perception and generalization there should 
be a very large variation in the behaviour of children in the generaliza- 
tion tests. 


In this question we return to the form of reasoning and the content 
to be structured ; the faculty of abstraction does not correspond to an 
operation which would appear at seven years whatever the object or 
the situation. Although a child of seven years may manage to classify 
red boxes and oppose them to the class of blue boxes, and may be 
able to classify big boxes and oppose them to the class of httle boxes, 
I cannot tell without previous experience whether he would be able 
as easily to form a class of primroses and to oppose it to the class of 
non-primrose flowers. In the same way the hierarchical pigeon- 
holing of animal classes (in our experiment it is a matter of ducks, 
birds, and animals) can be greatly facilitated when we oppose a 
non-animal class (chairs, books, etc.) to the animal class. 

The moment when the classification operations appear depends 
not only on the operational capacity as such but also, though in a 
lesser degree, on the content on which it operates. Obviously this is 
only a partial reply to your question. 


Dr. Lorenz spoke of a case where the common Gestalt encourages 
abstraction, whereas in most of our tests we try to put the perceptive 
configuration in conflict with the operation. That is why we have 
difiiculty in replying to you. 


Yes, in order to reply to my question it would be necessary to 
carry out tests where the Gestalt is absolutely useless. 


Without wishing to detract from your daughter I would say that 
a common Gestalt encourages abstraction and then we are not certain 
of being in the presence of what we call an operation. 


It can be shown that a machine or animal that can extract invari- 
ants in the form of Gestalten, or patterns, is automatically capable 
of learning. A machine or animal that has been designed, or has 


evolved, so that it can learn can automatically at the same time 
extract Gestalten. This is a first principle of cybernetic design which 
has been developed only quite recently as a theorem. 


I am very excited about this cybernetic fact. Learning facts hke 
that makes it worth while attending conferences! I may somewhat 
lessen my excitement by stating that Wolfgang Kohler (1940) 
knew one side of this and said that no animal can acquire knowledge 
of a stimulus situation as a signal releasing a conditioned reaction 
if it is not within the scope of its Gestalt perception. This works only 
one way, though, and it seems tremendously important if it can be 
shown that, conversely, a machine that can perceive Gestalt is auto- 
matically able to learn. 


This is a very good illustration, I think, of an intervention of 
cultural factors. The two illustrations of Dr. Lorenz are perfect 
examples. In the one case you have the small child who recognizes 
a mammal, and his parents are distressed, so they say 'No, that's 
a dog, bow-wow here', and they break the abstraction down to 
a concrete level. On the other hand, Dr. Lorenz's child, having grown 
up in a home where abstraction was valued, does the same thing and 
it is valued. We have a good many instances of this happening. The 
fourth word my child said was an abstraction. She said the word 
'baby' for her reflection, a doll, a photograph, and a carving. Because 
we were interested in linguistics and were interested in concept forma- 
tion we saw what she had done and we said it back, so it had some 
chance. So in your concrete cultures, in languages where there is a 
word for 'snow in the air' and another for 'snow on the ground' and 
another for 'snow that's crisp' and another for 'snow that's soft', 
undoubtedly the child at two sees snow and says 'That's snow' and is 
told 'No, you don't see snow, there's no such thing, it's this thing 
and this thing and this thing'. Then, when we come to the adult 
in various societies, we see one people very able to abstract and 
another people very handicapped in that respect. 


I think culture influences Gestalt perceptions even more because 
you have to look at something very often until the Gestalt jumps out 
at you. For example, the regularities contained in Eastern music are 
imperceptible to our Gestalt perception which is attuned to the 
eight-tone octave and won't respond to the scale of the Asiatics, and 


to us Asiatic music seems at first to be just chaotic. I doubt very 
strongly whether a grown-up person who is not particularly musically 
gifted is able to learn to feel the Gestalt and thus the beauty of 
Chinese music, and it would be an interesting experiment for Gestalt 
psychologists to see at what age you are still plastic enough to learn 
to do so. 


Let me come back to the theoretical point which you raised. The 
people who appear to us to see least are the Austrahan Aborigines, 
who apparently relate themselves to their environment much more 
by touch and by smell than by seeing. Australia has an extra- 
ordinarily monotonous landscape. The drawings of the Aborigines 
are always diagrammatic; they are not pictorial representations. But 
take Aboriginal children now and show them perspective and they 
get drunk. They get into an absolute ecstasy when they first encounter 
perspective, and small ten-year-olds will fight up and say 'I see, you've 
painted it the way it looks, not the way it really is'. 


It seems to me that in our Western European world the reality is 
essentially one of spatial and mechanical relations. The most striking 
thing one sees in Africans who have lived in the country and are not 
famihar with a European background is their utter incapability of 
dealing with spatial relations. I have tried to test Africans with blocks 
of wood, coloured on different sides : a cube-imitation test, a very 
simple thing for Europeans, even perhaps at about the age of eight; 
but the African adult, on average, is quite incapable of deafing with 
this, and I finally came to the conclusion that most of the people I 
examined did not reafize that, when they turned a cube round, the 
various sides maintained a constant relationship to each other. This 
is for us a very simple thing, and these Africans were in many ways 
intelligent, capable of deafing with their own environment extremely 
well, but in this skifi, which is acquired by us I suppose in playing 
with toys, they are feeble-minded by our standards. Reality for them 
is not a world of spatial relations, but a world of spirits, and our 
mechanical world is much less real to them. 


I think it is important to distinguish between the word 'reafity' as 
it is usually used in psycho-analysis or therapeutically, when you say 
you have to face reality — which includes your mother-in-law, the 
marriage systems, the secret pofice, or whatever happens to be 


around in your country — and reality in the sense that Dr. Lorenz 
was using it, when he said that the function of this mechanism was 
to inform the animal of the actual state of the object. Unless we 
keep those two things absolutely clear, we are going to get into a 
great deal of trouble. To date, I have seen no evidence anywhere of 
any people with an inability to interpret whether an object is moving 
or not (that is to function as Dr. Lorenz was describing) unless 
superimposed there is a theory of spirits, which then gives a new 
interpretation. But this is quite different from the original capacity 
to perceive. My interpretation, on the basis of the testing I have done 
with different primitive people, would be that there is something in 
Dr. Carothers' test which evokes an alternative interpretation in the 
African. After all, when the African turns his boat around he knows 
all about the sides. When he is dealing with any object in his own 
environment, he knows just what he appears not to know in that 
test situation. We have to think of culture as mediating between the 
actual nature of the object, reported in Dr. Lorenz's sense, and the 
final interpretation as to whether it is a ghost or a devil, or whether 
it was there or wasn't there. 


I think it is true to say that hearing plays a much more vivid part 
in the African's life than vision. He is familiar with the sounds of the 
wind in the trees, the running of water, the noises of the animals, 
and the speech of his relations and his friends. He is not nearly so 
insistently introduced to visual aspects of the world as European 
children are — reading, writing, playing with toys and blocks of 
wood. It seems to me that all these auditory stimuli are of a very 
emotional type. They are always relevant to the person. No matter 
what sound one hears it is always something that is of immediate 
interest, and even the human word as spoken is far more emotional, 
and of far more intense personal interest, than the written word. That 
is likely to play a part in the African's approach to life and to 
emphasize his subjectivity and his tendency to egotism and his lack 
of detachment. The visual world is also far more continuous and 
always reminds one of the fact that each event depends on previous 
events, that things do not happen haphazardly; whereas in the 
world of sound things are discontinuous and cause and effect are 
far less obvious. 


From the discussion which has just taken place I understood that 
everyone agreed that according to the different social or cultural 


environment variations in mental development occur as a function of 
education, emotional forces, etc. Nevertheless it seemed to me that 
Dr. Mead and Dr. Lorenz postulated a sort of constancy of Gestalt 
in the immediate sensorial universe, throughout the world. I per- 
sonally have some doubt on the subject. I do not think that all the 
Gestalten are the same for both child and adult. I will give an 
example from the constancies referred to by Dr. Lorenz when he was 
explaining his ideas on Gestalt. 

We carried out some research on the depth evaluation of real and 
apparent (projective) size. As regards Euclidian or real size we found 
that the constancy was lower in the child — as was demonstrated long 
ago by the psychologists of the Vienna school. The child undervalues 
depth. At about ten or eleven years he attains approximately the 
level of real constancy. The adult overestimates depth. There is an 
over-constancy in the adult and this over-constancy is the rule. Many 
psychologists had demonstrated this, but with a certain constraint, 
passing over this phenomenon like a cat on hot bricks because it 
cannot be reconciled with Gestalt. 

Now if apparent size is studied, that is, the way an object at four 
yards appears compared with an object at one yard, for instance, it is 
found that the adult has considerable difficulty in evaluating this 
projective size. He makes an error of about 50 per cent, and even 
more. In the case of a child of about seven years one has great diffi- 
culty in getting him to understand the question, but once he has 
understood it his perceptive evaluation is better than the adult's. His 
estimate of the apparent size is in some cases 100 per cent. In adults 
we have only found one case where the evaluation is as good as the 
young child's, and he was a landscape painter. Apart from this 
exception adults have a very inferior power of evaluation compared 
with children. 

There is a large number of phenomena of this kind, which shows 
that Gestalt, propounded as a simple, universal explanation, appears 
to be a myth from the point of view of genetic psychology. Of course, 
I readily admit that these phenomena depend not only on age but also 
on culture, and I think that even from the sensorial point of view 
wide differences are found. 


Some of you probably know of the experiments of Ames (1952) 
and the room he has constructed so as to give an illusion of being 
rectangular when actually it is extremely distorted. You go in and 
then you are given a stick and told to touch first this part of it, then 
that part of it; and it takes a long time for the motor experience 


to correct this visual illusion. In this httle house there are two 
windows of the same size, so constructed that when you look at 
people looking through them one person looks almost twice the size 
of the other. If we took members of this group into this experimental 
situation, if there were two of you behind the windows and one of us 
looking, we would get the illusion completely and accept the fact that 
the whole situation had been set up for it. But with a child looking 
at its parents, or a wife looking at her husband, the distortion does 
not operate and they see them with the whole perception corrected, 
so that there is not only the deceptive constancy of the adult but the 
constancy that is given by the emotional situation, (We are not quite 
sure what husbands see in wives, because the experimenters have a 
theory which relates to dependence and they maintain their theory 
by not having husbands looking at wives.) 


If a husband sees his wife the right size then at the same moment 
he ought to see the house askew. 


I might relate an experience with a boy. He saw his father and 
mother coming up an avenue and he said to me, 'Isn't it funny, why 
is daddy so small and mummy so big ?' Actually the mother was an 
inch or two shorter than her husband. The child, who was passing 
through the Oedipus phase, had an emotional tendency to diminish 
the size, i.e. the value, of his father. I think this is an illusion, a kind 
of 'disgestaltism', frequently found in psychoanalysis. 

I would hke to make another remark in connexion with Mile 
Inhelder's very interesting communication and M. Piaget's remarks. 
Many of you commonly speak of tests, but I think a distinction 
could be made between an actual test and what we might call an 
inquiry. The way Mile Inhelder and her students go about getting 
children to undergo testing seems to me different from the cold and 
automatic way testers apply tests just as they are to all children 
without discrimination. As I understand it, Mile Inhelder prepares 
the ground; with each child she tries to create some relationship 
with an affective component by explaining things to him and playing 
with him. In psychoanalysis we would say that Mile Inhelder attempts 
to create a positive transference. Now I have ascertained that in 
certain cases a negative transference is estabhshed and this affects 
the result of the test. 

I will quote you an actual case which borders on the anecdotal. I 
had occasion to supervise the work of a child psychoanalyst. She was 


treating an anxious boy of seven or eight who had been tested with 
the communicating vessel test by one of the Institut Rousseau 
students. This test consists in showing a child two communicating 
vessels. One of the vessels is screened and the water-level is varied. 
The child is then asked various questions to reveal the extent of his 
acquisition of the concepts of dynamics, volume, and horizontaUty. 
The child gives all kinds of interesting replies of an animistic or 
magical type, saying that the water is growing, etc. Well, this child 
finished by telling the psychoanalyst that he had been careful not to 
say what he thought. He had imagined that the tester wanted to 
know whether he had wet his bed. 

I have seen another case where a test included filling flasks with 
water and emptying them. The httle boy was suspicious. He made a 
negative transference and rephed all wrong because he thought the 
young lady — as he called her — wanted to know whether he had 
masturbated. Naturally in this case the result was falsified. 

These examples clearly show that two different factors must be 
distinguished in these tests and that a so-called intellectual test must 
not have an affective base, and inversely an affective test must not be 
carried out on an intellectual plane. 


I want to ask Mile Inhelder if she envisages these steps in mental 
development as definite steps with a pause in between, or does she 
envisage the mental development as continuous and these steps 
simply as milestones of measurement ? 


The evolution appears to be continuous; however, in this con- 
tinuous process there seem to be decisive periods: for example, at 
about seven years, which appear as accelerations. It all happens as if 
a slow preparation suddenly culminated in an achievement. 



Comparative Behaviourology 


If you pull a chair away from a man sitting on it and balancing 
crockery, he will put out his hands and let the crockery drop. This is 
due to a reflex. I am convinced that the thing called a reflex exists, 
but I am also convinced that it is very far indeed from being the only 
basic function of the central nervous system, as many physiologists 
still beheve. Someone once said that today's truth is tomorrow's 
error. Otto Koehler very wittily and very profoundly replied, 'No, 
today's truth is tomorrow's special case'. 

Most accepted 'truths' about the reflex are such special cases. None 
of us knows what is really happening within the central nervous 
system. Permit me to talk in parables. Supposing there is an auto- 
matic machine on the railway station. You put in a coin, then there is 
a mysterious buzz and out pops a package of cigarettes. The machine 
is the central nervous system, the coin is the stimulus and the 
cigarettes are behaviour. The buzz represents everything recorded by 
electrophysiology, and is the only hint we get as to what may be 
happening inside. 

Now let us, for the moment, forget all we think we know about the 
approved 'truths' concerning the reflex. Let us regard the central 
nervous system as such an automatic machine whose insides are quite 
unknown to us. Let us study its different types of performance, its 
different types of reaction to stimulation, and also its spontaneous 
activities, and let us classify them from the point of view of the 
stimulation impinging on the central nervous system and of the 
activities elicited by these stimuli. One of the most common types of 
central nervous performance is as follows. A continuous and rather 
amorphous stream of excitation comes in through aff"erent nerves and 
the output is a very well co-ordinated rhythmic sequence of motor 
impulses. Many instinctive movements, particularly those of loco- 
motion, belong to this type. Or one single impulse is sent in and a 
series of rhythmic impulses constitute the response as, for instance. 


in the case of a clonus. Among literally hundreds of such combina- 
tions between the impinging stimulation and the resulting efferent 
impulses, there is one case in which there is a direct correspondence 
between stimulation and efferent impulse, that is to say one stimulus 
elicits one efferent impulse, thus creating the extremely misleading 
impression that the central nervous system has not done anything but 
pass on a stimulus from one neuron to another; the prototype of what 
is called a reflex. This kind of central nervous response is neither 
common nor widely spread in the zoological system. On the contrary, 
the most typical and classical form of reflex, on which the whole 
conception was based, has evolved very late in phylogeny, as it only 
exists in animals with a pyramidal system. 

Once we have recognized it as one special case amongst hundreds 
of different ones, there is not the faintest reason to believe that all 
central nervous performances are based on 'the reflex' as a primary 
functional element; yet with some physiologists, you will find it 
extremely difficult to overcome this opinion! 

If we put the typical reflex — one stimulus, one efferent impulse — 
at one end of the diagram which classifies central nervous perfor- 
mances by their input and output, we find, at its other end, a number 
of functions which are independent of impinging afferent stimulation. 
This endogenous or automatic production of impulses may result in 
the output of a continuous stream of simply amorphous excitation, 
as has been shown by Holst (1939, Schoen and Holst 1950) for the 
sensory cells in the macula utriculi in the labyrinth of fishes, and 
termed 'autostaxis'. Or it may result in sending out very complicated 
and highly co-ordinated rhythmical series of motor impulses. 
Hoist not only deafferented, but totally isolated the central nervous 
system of the earthworm, and found that it still persisted in sending 
out rhythmical co-ordinated impulses. These impulses proceed in 
waves from the rear part of the front end and are indubitably identical 
with those normally eliciting the creeping movement of the worm. 
When this was doubted by some critics, Hoist proceeded to isolate 
only part of the nervous system and to leave part of it in connexion 
with several segments of the worm. Then the action currents con- 
ducted from the isolated part to a series of galvanometers went 
beautifully and convincingly in step with the contractions of the intact 
segments. This type of central nervous activity, producing, without 
any afferent stimulation or control, rhythmical and co-ordinated 
movements, was called 'autorhythmia' by Hoist. Pure autorhythmia 
is, in the animal kingdom, about as rare as pure reflex activity, but it 
is indubitably a much more primitive phenomenon. It is very prob- 
able that all locomotion in protozoans is closely akin to auto- 
rhythmia, and Bethe has shown conclusively that the contractions of 


the umbrella of some jellyfish are autorhythmic. But in some more 
highly differentiated types of jellyfish, the Scyphomedusae, the auto- 
rhythmia of the velum has been graduUy superseded by a typical 
reflex mechanism. These animals developed static organs called 
marginal bodies which primarily served a righting response by 
stimulating a stronger contraction on the lower rim of the umbrella 
whenever the animal was tilted to one side. But gradually, in a manner 
which need not concern us here, the reflex mechanism of this righting 
response has replaced and superseded the autorhythmia of the velum. 
Thus the jellyfishes, meek animals though they are, may serve as a 
warning to the physiologist never to generalize rashly. The umbrella- 
ing movement of a primitive Hydromedusa not only looks exactly 
like that of a Scyphomedusa, but actually is the 'same' movement 
ontogenetically and phylogenetically. Yet the one is caused by the 
purest type of autorhythmia that we know and the second by the 
purest type of reflex that we know. I confess, however, that it gives 
me some satisfaction to remind you of the fact that autorhythmia is 
certainly the more primitive and the older of the two. 

I deemed this excursion into neurophysiology necessary in order 
to explain why ethologists absolutely refuse to accept the current 
explanation that all instinctive activities are 'chains of reflexes'. No 
complex instinctive movement has, as yet, been really analysed down 
to its neurological components. Yet there is overwhelming evidence 
that true autorhythmia plays the most important part in the causation 
of the typical spontaneity of all instinctive activity. This assumption 
supplies us with a most convincing explanation of a very great number 
of facts concerning instinctive movement, while the same facts remain 
totally unintelhgible on the basis of the chain reflex theory. And the 
fact that all instinctive activities are fundamentally spontaneous and 
therefore akin to autorhythmia is relevant for what I want to say to 
you about what we call the innate releasing mechanism (I.R.M.). 
As long as the elaborate and well co-ordinated sequence of movements 
constituting an instinctive activity was regarded as a chain of reflexes, 
its beginning did not obtrude itself as a problem: it was just the first 
one among other reflexes and the existence of a neural mechanism of 
a peculiar character did not become apparent. But as soon as it 
became evident that the elementary nervous functions underlying 
instinctive movements were much more akin to autorhythmia than to 
reflex activity, a new problem put itself very directly. If the instinc- 
tive activity is generating impulses continuously — as there is plenty 
of evidence that it does — and prevented from causing incessant 
movement only by the inhibiting function of higher centres, how is 
this inhibition removed at the right moment and in the biologically 
adequate situation in which the instinctive activity must be 


discharged? We know that the ventral chord of the earthworm 
generates the impulses of creeping incessantly, yet the intact earth- 
worm creeps only when it ought to. 

There is no contradicting the assertion that the 'trigger-mechanism' 
removing the central inhibition which, most of the time, prevents 
instinctive activities from 'going off in vacuo' is nothing else than an 
'unconditioned reflex' in the original sense of I. P. Pavlov. Yet this 
assertion does not solve the real problem, which actually Ues in the 
selectivity of the releasing mechanism. This central problem only 
becomes apparent in the more complicated forms of the I.R.M., not 
in the simple cases that Pavlov's classical examples provide. When a 
dog secretes two different kinds of saliva on having meat extract put 
in his mouth in one case and sand in the other, the difference between 
the two reactions is easily explained by the fact that the meat extract 
stimulates chemical receptors and the sand tactile ones. But take the 
following case as a contrast. A young totally inexperienced jackdaw 
instantly reacts by doing 'social grooming', i.e. preening the other's 
feathers, if another assumes the 'friendly' or submissive attitude which 
consists in turning away the beak and fluffing the feathers at the back 
of the head. But if the other bird instead of fluffing the feathers of the 
head fluffs those of the back, and instead of turning the bill away 
turns it towards the approaching youngster, the latter will instantly 
'recognize' this attitude for that of a threat, and react either by 
retreating or by a counter-threat, according to its relative strength and 
courage. To anybody not beheving in vitahstic miracles, these two 
different ways of reacting to two stimulus situations which, after all, 
are only different in regard to the retinal images received of the sub- 
missive and the aggressive jackdaw, must be a source of deep wonder. 
We have to postulate an innate perceptual structure which acts as a 
sort of filter letting through only sharply defined combinations of 
sensory data. Thus the problem does not he in the question of whether 
this performance is due to an unconditioned response or not; it lies, 
so to say, in the afferent side of the reflex arc. 

If we do not know what an innate releasing mechanism (I.R.M.) 
really is, we know quite a lot already about its function. A lot of 
investigators, chiefly Tinbergen and his school, have concentrated on 
the I.R.M., so that at present it really is the best known element of 
innate behaviour, I cannot give you here even a short survey of 
everything we think we have brought to fight about the function of 
the I.R.M. ; I only want to emphasize a few points. 

{A) An I.R.M. never responds selectively to a complicated Gestalt, 
but exclusively to extremely simple key stimufi. These may be 
relational properties, but always such simple ones that the relation in 
question can be stated in very few words. For example, the I.R.M. 


which, in the young herring gull, elicits begging activity, responds to 
the following stimuli emanating from the parent bird's bill: 

(a) The narrowness of the lower mandible which, in feeding, is 
turned with its sharp under edge towards the baby bird. 

(b) The red colour of a round spot on the lower mandible of the 
parent bird. 

(c) The contrast of this spot against the paler coloration of the 

(d) The downward tilt of the bill. 

(e) Its jerking movements while the parent bird is regurgitating 

The general form and colour pattern of the parent's head, bill, 
eyes, etc., are quite irrelevant for the releasing power of the situation. 

(B) If, as in most cases, an I.R.M. responds to more than one key- 
stimulus, the effect of stimulation is exactly in proportion to the sum 
of the efficacies of the several key stimuli impinging at the moment. 
This rule, termed the 'law of heterogeneous summation', constitutes 
a very striking difference between the learned response to Gestalt 
qualities and the innate reaction of key stimuli. Furthermore, I want 
to draw your attention to the fact that this very limitation of an 
I.R.M.'s functions, particularly the limitation of their complexity, is 
just what one would postulate from a purely mechanistic, or, if you 
like, cybernetic point of view. It is quite to be expected that a 
'receiving set' reacting selectively to a few and very pregnant key 
stimuli is easier to construct than one which reacts to a diffuse com- 
plex quality of an elaborate Gestalt. Indeed, the very simplicity of the 
I.R.M. is a reason for analytical optimism! Though our ideas about 
what the I.R.M. really may be neurophysiologically are purely 
conjectural, I think that the construction of a cybernetic machine of 
similar functions and Hmitations would be very illuminating. 

As our ideas about that peculiar stimulus-selecting apparatus 
which we call the I.R.M. are largely founded on our observations 
about what it cannot do, it will not seem surprising to you that the 
pathology of the I.R.M. is also one of the major sources of our 
knowledge about it. After all it is the approved method of neuro- 
physiology to study the pathological defects of a function in order to 
gain insight into its normal causal structure. Another reason for 
enlarging a bit on the pathology of the I.R.M. on this occasion lies in 
the fact that it might have some bearing on some phenomena of 
delinquency in human social behaviour. 

One basic fact which is extremely characteristic of the normal 
function of all I.R.M.s must be stated first of all: in all cases in which 
a pathological disintegration of an I.R.M. is found, the activity 
normally released by that I.R.M. does not, by any means, become 


unreleasable, but, on the contrary, more easy to release. This fact 
very strongly enhances the I.R.M.'s character of a 'filter' of stimuli: 
the more that 'filter' is broken up, the greater becomes the range of 
stimuli which are able to pass through it and to release the reaction 
in question; in other words, the less becomes the selectivity of the 

There are three different factors which regularly lead to a disintegra- 
tion of the I.R.M. and to a corresponding loss of its selectivity: 

(A) Any, even a very slight, disturbance in the general state of 
health of an animal. 

(B) The domestication of a species. 

(C) The hybridization of two species with shghtly different 

Let me first give you an example of (A). The red-backed shrike 
{Lanius collurio L.) has an instinctive movement to impale insects and 
other prey on thorns, in order to store food. In young birds of that 
species, reared by me in isolation, I found that the innate movements 
of impahng were not innately directed towards a thorn. They tried to 
impale a mealworm or a small piece of meat indiscriminately every- 
where; on their perches, on the bars of their cage, etc., without giving 
the slightest attention to the very adequate artificial thorns with 
which I had supphed them. Only when, by pure trial and error, they 
happened to execute the innate movement pattern of impaling on one 
of these thorns, the full success of the instinctive movement evidently 
acted as a reinforcement to direct the impaling activity to the adequate 
object, and they learned with extreme rapidity. Naturally, my con- 
clusion was that the red-backed shrike had no I.R.M. directing its 
impaling activity to the thorn, but that it had to learn its proper use 
by individual experience. This, however, proved to be quite erroneous. 
Kramer, of Wilhelmshaven, reared young shrikes in order to study 
their migration activities. Just because an easy opportunity offered 
itself, he repeated my experiments on the impaling activity — with 
exactly the same results. But one year later he improved his rearing 
technique, feeding the young birds on a large proportion of live silk- 
worms. And when he repeated our experiments with these birds 
which were in just slightly better condition than all those previously 
used, he found that they had an innate reaction to the thorn ! At the 
very first experiment they took the mealworm in their bill, looked 
about for an adequate thorn, recognized it instantly when they saw 
one, went straight for it and, without the least evidence of trial and 
error, impaled their worm on the thorn, just as if they had done so 
hundreds of times. 

Now an example of (5). In the wild ancestor of our domestic 
chicken, the jungle fowl or bankiva, the mother hen refuses to 


brood any chick which has not got the typical wild colour pattern 
of down on its head and its back. Some bankiva hens instantly kill 
any black chick. In domestic hens one finds all possible gradations 
between this extreme selectivity of the I.R.M. releasing the brooding 
activities, and an entire lack of selectivity. Most barnyard hens are 
insusceptible to all the possible colour patterns in chicks, but some 
will refuse to brood duckhngs or goshngs ; highbred races, like cochins 
or brahmas, will brood practically anything aHve and approximately 
of the right size — young ferrets, for instance. 

Now to come to (C), the loss of selectivity of the I.R.M. in 
hybrids : some hybrids of bahama pintail and chestnut-breasted teal 
that had been reared normally, in company with their brothers and 
sisters, did not react sexually either to each other or to any birds of 
the two parent species, but all of them, both males and females, 
courted a huge white-spotted domestic duck, about four times their 
own size. Similarly, Heinroth found that some goose hybrids persisted 
in courting swans. Though these are just a few isolated observations, 
they tend to show that loss of selectivity of the I.R.M. stimuli may 
result in the choice of the strongest stimuli available. 

Finally I want to mention one peculiarity of the I.R.M. which, 
though not pathological in itself, may lead to phenomena closely 
akin to the pathological. In most I.R.M.s, the stimulus-receiving set 
is tuned to the quality rather than to the quantity and intensity of 
the natural key stimuli. Therefore most key stimuli can be exagger- 
ated. In the jewel fish, Hemischromis bimaculatus, the dark ruby-red 
colouring of the male's throat is one of the key stimuli releasing 
fighting activities in a rival. By illuminating our fighting arena with 
ruby light, thus intensifying their colour, we can make jewel fish 
males 'see red' and fight each other more intensely than under normal 
conditions. If more than one of the key stimuli emanating from a 
certain object can be thus exaggerated, it is possible to construct a 
model which by far surpasses the releasing effect of the natural 
object. In the oyster-catcher, for instance, the I.R.M. which responds 
to the egg and elicits incubation activities is dependent, among other 
key stimuli, on the size, the colour, and the contrasting spots of the 
egg. A real oyster-catcher egg is the same size as a bantam's, its 
colour is bluish grey with slightly darker grey spots. If one presents 
an incubating oyster-catcher with an egg nearly as large as an 
ostrich's, of bright blue colour and with deep black spots, thus 
exaggerating the intensity of the key stimuli mentioned, the bird 
becomes absolutely fascinated, leaves its own clutch and passionately 
tries to incubate the giant egg, though this is physically impossible, 
the bird being hardly able even to stand astride the model. A very 
intelligent American journalist, on seeing Tinbergen's film showing 


this behaviour, exclaimed: 'Why, that's the cover-girl!' This witty 
remark is scientifically quite correct. Much of what we call 'vice' in 
human behaviour does not consist in anything else but the search for 
supra-normal key stimuli. The vice of gluttony offers very convincing 
examples, and quite proper ones at that. 


There is one particular type of I.R.M. whose function is closely 
linked with a particular type of conditioning. I think it may be of 
interest to you, because this interaction between I.R.M. and learning 
is limited to a strictly defined phase in the organism's ontogenetic 
development. These I.R.M. s are of an extreme simplicity and there- 
fore their selectivity is shght. But this lack of selectivity is compen- 
sated by the Hmitation of the time during which the I.R.M. is effective. 
During that short period, the I.R.M. succeeds, under natural 
circumstances, in establishing a conditioned response to its object. 
The resulting response then is far more selective than any I.R.M., 
as all learned responses always are. Let me give an example. The 
I.R.M.s ehciting, in a newly hatched greylag gosling, the activity of 
following its mother respond to an amazingly wide range of key 
stimuli. Any object between the sizes of a bantam hen and a big 
row-boat, which moves and emits noises of a wide variation of pitch, 
can release the following response of a newborn greylag. If a man 
moves and talks in the presence of such a little gosling, the latter will 
look at him very intently, give its greeting response, and, after a few 
repetitions, follow him unconditionally, just as it would normally 
follow its mother. Obviously this combination of a simple I.R.M. and 
consequent conditioning is entirely effective under natural conditions. 
The cases in which the mother goose is not the first moving and 
sound-emitting object perceived by the gosling are, of course, so 
extremely rare as to be of no account. 

Now the particular kind of conditioning that takes place in the 
process just described differs from other types of learning in a number 
of very characteristic points which I want to summarize : 

{a) It is limited to a very definite and often extremely short phase 
of ontogeny. 

{b) It is, in the typical cases, quite irreversible. 

(c) It takes place quite independently of whether the activity 
released by the stimulus situation is, at the time being, functional 
or not. 

Mainly because of its irreversibility we have called this particular 
type of conditioning imprinting. The last of these three points is very 
important, because you cannot get any other conditioned response 
when the unconditioned response is not yet functional. That is what 


makes us believe that imprinting is something which takes place in 
the perceptive sector only, because the efferent sector of the reaction- 
arc need not be present at all. 

There are two points about imprinting which seem important in 
this connexion. First I want to call your attention to a very enigmatic 
fact. The gosling in the classical imprinting experiment of Heinroth 
does not become imprinted to the particular man whom it sees at 
first, but to Man, with a capital, as a species. It can learn to know its 
keeper, later on, by a common learning process, but the irreversible 
imprint refers to the species and not the individual. We have no 
explanation of this at all. 

Secondly, I want to emphasize that most probably there are all 
imaginable types of conditioning, forming gradations between true 
imprinting and the more common types of learning. What we call 
imprinting is a type we found at first in jackdaws and greylag geese. 
It is one extreme type of learning, and there may be all sorts of grada- 
tions. We know already from the work of Eckhardt Hess that, 
for instance, depth perception is acquired in a way very similar 
to imprinting, and if you prevent a chick from acquiring it during 
the first days, it cannot do so afterwards. That is not a fixation 
of a simple activity to its objects, but it is like imprinting in being 
confined to a very short phase of the ontogenetic development of the 
individual. I mention this only in order to emphasize that there may 
be all sorts of gradations ; that is one example of extreme imprinting 
in some respects and not in others. There may be a superposition and 
interlocking of imprinting and learning. Such intercalations, though 
superficially tending to veil the irreversibility of imprinting, really 
afford its most convincing proof. The budgerigar, to give you a good 
example of this, is a bird which learns easily to accept substitute 
objects for its sexual activities. You have probably seen budgerigars 
in cages court a celluloid doll, etc. We imprinted the sexual activities 
of two budgerigars, a male and a female, to the human species — 
which it is quite easy to do by rearing the bird in isolation during a 
certain period. Then we deprived them of all human company and 
kept them together in a lonely garret where they were fed and 
watered through chutes. After a time they learned to use each other 
for substitute objects, courted and copulated quite normally and 
even reared two broods with full success. Now after the third brood 
had hatched, we did the crucial experiment which simply consisted 
in my going into that garret. Both birds instantly went into a frenzy 
of sexual excitement which was directed entirely towards my person. 
The impression of this reunion with what the birds 'considered' their 
own species was so lasting that they absolutely refused to have any- 
thing to do with each other for a long time and let their brood of 


young die of hunger. Now consider that these birds had many times 
successfully copulated with each other, while all their attempts to 
copulate with the imprinted object of their passion necessarily had 
always remained unsuccessful and unsatisfying. Nevertheless they 
persisted in their object-fixation, if I may call it thus! Though 
imprinting has been found in its typical form in birds and insects 
rather than in mammals, I really do believe it to be fundamentally 
akin to those very lasting object-fixations of human beings, chiefly 
because these fixations also seem to be dependent on early childhood 
impressions and seem also to be largely irreversible. Some psychia- 
trists and psychoanalysts here I believe share this opinion, at least 
as a working hypothesis. 

I now will proceed to show you a short film illustrating some of my 
points. The first part of the film shows the behaviour of a flock of 
young goslings imprinted to man, in comparison with another flock, 
imprinted to its own species, and the second part shows reproductive 

Dr. Lorenz' film illustrated the following response, reaction to 
warning call, unmixing of the two differently imprinted flocks, 
following response on the wing, greeting reaction, pair formation, 
copulation activities, and neck-dipping. 


Dr. Lorenz started his introduction by speaking of reflexes and 
seemed to say that actually reflexes do not exist. . . . 


No, excuse me. I said that the reflex is a special case of response 
of the central nervous system, which exists in its classical form in 
certain animals capable of a high degree of differentiation. 


How would you call the retracting movement of an oyster when 
touched with the prongs of a fork ? 


That is just the point : I would not give it a name at all, before 
knowing what it is physiologically! All functions of the central 
nervous system ought to be termed in a way containing as little 
hypothesis as possible. The thing to do is to give a purely descriptive 
term corresponding to what we know about incoming stimulation 


and outgoing activity. In your example of the oyster the input is 
very probably a single wave of excitation and the output consists 
in a long-lasting increase of muscular tonus — most muscular con- 
tractions of molluscs are tonic. This physiological process is certainly 
'a reaction', but very different indeed from a 'rejQex'. This word always 
calls up the mental picture of the classic diagram of the cross section 
of the cord with the short reflex arc coming in at the posterior root 
and going out by the anterior root. 

Then you limit the definition of reflex to this set-up? 


More exactly to the function of this set-up, which consists of a 
single wave of elferent excitation responding to a single wave of 
afferent excitation. This is certainly a reflex, but it is very rare in 
animal behaviour. The only examples I can give are those I have 
already quoted. Autorhythmia, as in the creeping movement of the 
earthworm, is as rare as the other, and between these two extremes 
there are thousands of kinds of function of the nervous system. 


You spoke of autorhythmia as if it were exceptional, but is not 
this property in fact one of the most general of the nervous system, 
at least from the electroencephalographic point of view? It has been 
found in the most elementary structures. Thus, it was demonstrated 
long ago by Adrian (1931) in the optic ganghon of the Dystkus 
marginalis. He was able to show that even when isolated this ganglion 
was in continuous activity. When it is connected to the eye, or what 
replaces the eye in this animal, this activity ceases immediately light 
arrives. Autorhythmia is a kind of negative to activity. 


It is the lack of electrical activity, of physiological activity, isn't it ? 


Certainly. That is exactly what I was coming to. 


I actually had intended to cite the sensory cells of the mucous 
membrane of the olfactory tract as an example, whose particular 


function also is something extremely queer and rare. When the out- 
put of the central nervous system is arhythmic but continues inde- 
pendently of afferent excitation, this function is termed 'autostaxis'. 
This autostaxis of the olfactory sensory cells can change into different 
autorhythmias, according to the smell which impinges on it. The 
moment a smell impinges on these olfactory cells, they begin to fire 
rhythmically in variable rhythms and each of these different rhythms 
is characteristic of one particular smell, Adrian can look at the curve 
of electrical activity given out by those cells and is able to say what 
kind of smell is impinging on the mucous membrane of the rabbit 
(Adrian and Ludw^g, 1938). That is an immense achievement. Of 
course, the reflex on the one hand, and the autostaxis and auto- 
rhythmias on the other hand, are joined to each other by a number of 
gradations: for instance, my pupil Prechtl (1952b) showed that the 
gripping reflex of the human child is apt to be a rhythmic movement 
in the first postnatal period. After stimulation it occurs several times. 
Only with the growth of the afferent control, the unnecessary and 
even detrimental sequence of movements is cut out and one single 
movement is left. The same is true of what Prechtl and Schleidt 
(1951) call the 'search automatisms' in young mammals and human 
babies. These have an autorhythmic movement of searching for 
the teat by turning their head to and fro. Now if you analyse this 
function, you find a very exceptional connexion between an auto- 
matism and an I.R.M. In the typical cases, the automatism is blocked 
all the time, until released by an I.R.M. which removes the block. 
Here, it is the other way round. The searching automatism is going on 
all the time, as long as the little animal is awake. It only ceases when 
a very definite stimulus situation is attained. In the cat, the respon- 
sible key stimuli emanate from the hairless area round the mother's 
nipple. As soon as the kitten finds this, its searching automatism is 
blocked. In other words, the I.R.M. puts on a block, instead of 
removing it as it does in most other cases. In the kitten, as well as 
in the human baby, it often happens that the I.R.M. responding to 
the nipple fails to inhibit the to and fro movement of the head at 
once, but does so with a slight delay, so that the little creature 
seems to look for the teat in the wrong place, which gives an effect 
very similar to the superposition of flight reaction and prehensile 
reaction which we saw in Dr. Monnier's film. 

What is called the 'prehensile reflex' in the human baby, is nothing 
other than the search automatism brought under a strict control of 
afferent functions. These release the automatism on a slight touch of 
the corner of the mouth, allow it just one sideways stroke, and stop 
it at once when the nipple is found. The ontogenetic development of 
the 'prehensile reflex' out of the 'search automatism' has been 


thoroughly demonstrated by Prechtl. And yet you may call it a reflex ! 
For very possibly a reflex, even in its purest form, is nothing but an 
autorhythmic process brought under the rein of afferent control to 
such an extent that it is just de-blocked, allowed to discharge just 
one stroke of excitation, and instantly blocked again. That, of course, 
is only theory. I only v^ant to repeat that I neither assert that 'there is 
no such thing as a reflex', nor do I believe that autorhythmia 'explains 
everything', but that there are lots of other types of central nervous 
functions, just as important as these tvv'O, and that it is sheer prejudice 
to assume that one of these many different performances is 'the' 
elementary function of the central nervous system. 

Now, apropos of the film, allow me just a few words on a curious 
phenomenon which we found in the reaction of the gosHngs to our 
imitation of the mother's warning call. It invariably was strongest at 
the first experiment and tended to fade very rapidly indeed with each 
repetition. Attempts to reinforce it were quite unsuccessful. The 
fading could not be prevented by letting even very strongly frightening 
stimuli impinge on the goslings immediately after the warning call. 
And even after a very long period of rest, the former intensity of the 
reaction does not become re-estabhshed. This is one of the chief 
problems which we are trying to solve at the moment. 


A similar reaction in man is an absolutely natural and physio- 
logical phenomenon which occurs even during sleep. If you take as a 
reference the existence of a cerebral electric reaction to stimulation 
during sleep, you find that on repetition of a known stimulation the 
response very rapidly weakens until it subsides completely. If the 
nature of the stimulus is modified somewhat the reaction reappears, 
but it fades away again. On returning to the first stimulation after a 
certain time, it is again found efiicacious. It seems that this is a very 
general property of the nervous system, which is on the alert only 
for what is new and unexpected; but as soon as there has been some 
kind of apprenticeship — and this apprenticeship does not necessitate 
a conscious vigilance — it economizes part or all of the reaction. This 
does not mean to say that vigilance is lost, since the least modification 
in stimulus causes reappearance of the reaction. 


I am very glad that Dr. Remond brought up this question, which I 
did not make clear. Of course, these are phenomena which the physio- 
logists have called adaptation; a rather unhappy term, because this 
type of adaptation has nothing to do with adaptation in the biological 


sense. Processes of physiological 'adaptation' may be biologically 
adaptive, or they may not. In order to make clear what I meant by 
special cases of 'adaptation', I had better describe the experiment of 
Robert Hinde (1953). Hinde found an I.R.M., whose effect he could 
quantify very well, the mobbing activity of the chaffinch. Chaffinches, 
hke other small birds, react to owls by giving their warning call, 
approaching the owl to a certain distance, and then following it about 
uttering their warning calls all the time, executing certain movements 
of characteristic excitement, actually displacement activities. The 
activity goes on for a measurable time, after which they quieten 
down. Hinde originally wanted to analyse the I.R.M. eliciting the 
mobbing reaction of the chaffinch, but what he discovered was some- 
thing much more important. The reaction depended on a very simple 
set of key stimuli, as most I.R.M.s do. The owl dummy only has to 
be round, it has to have eyes in a certain place, and so on. Now he 
quantified the number of of sideways movements and the number of 
'pink, pink, pink!' elicited by a standard dummy, until the reaction 
quietened down. The average of the number of sideways movements 
and the number of 'pinks' he found in virgin birds was taken as 
100 per cent. 

Then he found that in a second trial on the same day he got about 
30 per cent of that, and in a third trial he got about 10 per cent, and 
then, finally, 1 per cent. After a certain period of quiescence of one or 
several days he again got about 50 per cent and after a period of 
quiescence of six months he came up to 62 per cent, and that was all. 
Now we have no doubt that the reaction of the chaffinch to the owl 
has survival value. It was very surprising to us that this reaction is 
devalued by half if this unlucky bird happens to meet an owl once who 
who isn't hungry and is for several hours in that region. If this is so 
the reaction loses its biological usefulness altogether. Hinde thought 
that perhaps reinforcement was necessary. This was already a measure 
of desperation, because it is obvious that if the reaction was dependent 
on the bird getting nearly caught or actually pinched by the owl— 
which would be the only manner of reinforcing it — the reaction 
would be of little use to the bird. Well, he didn't succeed any more 
than I did in reinforcing the goslings' reaction to the warning call. 
We found that quite generally this 'fading' of reactions to I.R.M.s 
takes place whenever we elicit them with dummies. We strongly 
suspect that there must be something wrong with our experiments. 
What I propose to do now is this. Next year, I will take one of my 
greylag geese who is now two years old, and I will incubate and rear 
the young. Then I shall make the mother-goose warn, which I can 
easily do by presenting her with one of our eagles, or with a stuffed 
cat, or stuffed stoat, on a wire. The mother warns her flock twice. I 


shall have also a flock imprinted to me. I shall warn my flock exactly 
the number of times the mother does. I shall go on doing this and then 
we'll see if there is some difference in the fading between my goshngs 
and those of the mother, because what I suspect is that in some way, 
with this rather crude model, you supply a key which doesn't quite 
fit, and in some way spoils the lock. 

On the other hand, the phenomenon of adaptation is intriguing us 
in a very different manner. There is, in the reaction to the key stimuli 
constituting the I.R.M., a phenomenon of 'adaptation' which is most 
important. Prechtl (1953) did a simple experiment with chaffinches 
in their nest which have a gaping reaction that is very nearly a 
'reflex'. It can be elicited (a) by the slight trembling of the nest when 
the parent bird alights, and (b) by the call-note of the parent bird. 
These two stimuli are about equally effective quantitatively. Now you 
can stimulate gaping by shaking the nest and get about five or six 
reactions after which the bird doesn't react any more. Then you give 
the auditory stimulus and get six reactions more. If you invert the 
sequence of the stimuli, the numbers of responses remain virtually 
the same, which shows that neither of the stimuli is a priori more 
effective than the other, but that each of them becomes less effective 
with the number of repetitions. This 'blunting' of the reaction to one 
particular key stimulus (I prefer the word 'blunting' or, in German 
'Abstumpfung' to the ambiguous term 'adaptation') is a phenomenon 
which we have hitherto neglected. It is a very important source of 
error in all attempts to quantify the exhaustion of the efferent side, 
of the instinctive movement itself. Also, it is a source of error in the 
quantification of the efficacy of the single key stimuli, as compared 
with each other. One cannot test one key stimulus without blunting 
the response to it, and therefore any new stimulus, tested after some 
other, appears to be relatively stronger than it really is. Furthermore, 
there is one queer thing about this blunting or adaptation which we 
cannot explain at all: if, in the experiment with the gaping baby 
chaffinch, the reaction to one of the two key stimuh is totally 
exhausted before the second stimulus is brought into play, the total 
of responses elicited by both amounts to an average of eleven. But 
if, on the other hand, the change of stimuh is effected before the first 
one has become completely blunted, and then a second change, back 
to the first stimulus, is made before the second one has become 
ineffective, and so on, back and forth, a total of forty-eight reactions 
can be released on the average. 


Suppose that the chaffinches who could react to either of two 
stimuli got them both at once, what would happen then ? 



You would get a reaction of high intensity which might last con- 
siderably longer than the reaction to one of them. But you get a 
smaller number of reactions than by changing the stimuli. 


In the case of the gosHngs whose reaction to warning of the owl's 
presence appears to fade, is there any evidence that if you allow your 
real or model hawk to catch one of the goslings the fading becomes 
less ? In other words, does the experience of a catastrophe to a flock 
act as a reinforcement — does that tend to keep the response up or is 
there no such interaction ? 


We are not quite able to answer this. We did rather mild and not 
very thorough experiments by just letting the hawk swoop a bit and 
we didn't get any influence from that. Prechtl didn't get any influence 
by feeding his birds after gaping, which would have been a reinforce- 
ment. I think that when an instinctive movement does lead up to a 
definite consummatory act, then the consummatory act tends to act 
as a quite definite reinforcement ; there is no instance of fading in 
that type of I.R.M. Fading only occurs in another type of instinctive 
activity, in which there is no definite consummatory act. 

Dr. Remond spoke about a little change in the stimulus allowing 
the full reaction to occur again. In this regard Seitz (1940) could 
get the whole courtship activity of the male astatotilapia, up to 
fertilizing the egg, with dummy experiments. (The last step is 
elicited by olfactory stimulation emanating from the eggs, and he 
couldn't get that because the plasticine dummy didn't lay eggs.) When 
he painted the dummy a slightly different colour the reaction returned 
after fading and the curve of response rose quickly, but also was 
lowered more quickly. After repeated change of dummy the fish 
learnt that even a new dummy wouldn't be any good. He was 'dis- 
illusioned' as to dummies. But this 'disillusionment' is certainly not 
identical with fading. 


Bronk showed, if I remember correctly, that the response of an 
isolated sympathetic nervous ganglion to one type of input would 
depend upon whether or not it had received previously a different 
type of input along the same or along another nerve. One could even 
say of the isolated sympathetic ganghon that its response to a given 


stimulus depended upon its past history. It seems to me that there 
must be a continuing effect of some sort in these cases also, a residual 
that has been left in. 


The whole nervous system is designed to deal with a complex 
environment in which what matters is not that something particular 
occurs but that something is related to many other things. The 
isolated peripheral nerve fibre is a special case in which this is not 
true, because the isolated peripheral nerve fibre is purely a communi- 
cation system, not an analysing system; but the moment you get into 
the spinal cord, the importance of coincidence and the importance of 
time-relations between stimuH becomes obvious. 


Isn't that also true in the isolated nerve fibre ? 


It is not quite true, it's rather doubtful, I think. By mutilation you 
can make a nerve fibre regress to something a little bit more un- 
specialized; for instance, by cutting the ends and joining two together 
you can make an artificial synapse, but this is really a very special 
case of experimental interference. 


I should hke to mention the experiment of Birukov (1952), of 
Freiburg. He tried to exhaust righting reactions in frogs. Righting 
reactions have a tremendous amount of energy at their disposal and 
take an immense time to get tired. He found that on a wobbhng 
surface the righting reaction tired after several thousand attempts. 
When the frog was tired, and didn't right itself any more, a change of 
axis of tilt of only a few minutes of angle resulted in awakening the 
reaction again. Only then it got tired sooner, and the smaller the 
change in angle the sooner it tired. 


There are two questions I want to ask, the first regarding the 
notions of 'fading' and 'disillusionment'. Is it true that these are two 
separate processes ? You mentioned the 'fading' that can occur when 
I.R.M.s are elicited with dummies and you used the analogy of the 
key that doesn't quite fit and spoils the lock. In the second illustra- 
tion, when you used the word 'disillusionment', you said that the 


consummatory act reinforces the response and that a certain fish 
learnt that dummies were no good. If you take that fish back to a real 
female, are its copulatory activities diminished ? 


Well, we haven't done that, but to our isolated fish the female 
would have been nothing but another very new dummy and I should 
predict that he would react strongly and come to copulation with the 
female and this ultimate 'success' would reinforce the reactions to it. 
He would give the reaction again and again with her. But I agree that 
fading in I.R.M.-elicited activity without consummatory action is 
something different from the fading which is brought about by non- 
reinforcement, because in the cases in which the satisfaction of the 
consummatory act is lacking, the 'extinguishing' of the response is 
due to true learning. 


I should like to ask Dr. Lorenz whether the warning reaction given 
by the mother goose to her young when she sees the eagle depends on 
experience or whether it is absolutely spontaneous the fixst time an 
eagle goes by. And would the effect on the gosHngs be different if a 
duck or a goose went by instead of an eagle when the warning call is 


Well, I am afraid I must answer this question very extensively, in 
order to prevent generalization. I will answer it for the goose first. 
In the goose, the reaction to the eagle is purely innate, and the most 
curious thing is that the little geese look at the eagle when the mother 
is warning, but do not react to it afterwards on their own account. 
They do not learn by being warned, nor later remember it. It is not 
necessary, because they keep together with their mother. If you try 
to make experiments on a hand-reared goose it does not react to the 
eagle at all until it is fully fledged, and then it suddenly, at about nine 
weeks, begins to react to the eagle with an immense intensity, 
irrespective of whether it has seen eagles before. The I.R.M. of the 
eagle in the goose is extremely simple. An 'eagle' is anything which 
(a) is depicted against the sky as a background, (b) does not beat its 
wings, and (c) moves slowly. A goose will give the full eagle-reaction 
to a black feather drifting slowly in the breeze, or a pigeon gliding 
slowly against a strong head wind without beating its wings. You see 
all the geese giving full eagle-reaction to the pigeon, and when the 
pigeon gives a few wing-beats all the geese give an immense relief 


displacement-shaking and are quiet again. One funny thing takes 
place, which is just the opposite of imprinting. My first goose, 
Martina, had no experience of being warned by parents and didn't 
react to aeroplanes which were flying there all the time. She did not 
seem to notice the aeroplanes until her I.R.M. was mature, and then 
she suddenly got awfully afraid of them. But after a time she quietened 
down again, and got adapted to the aeroplanes flying over her. 

Now let's take another bird, the jackdaw. It has no innate reaction 
to enemies at all. One single case of hearing the warning call of the 
mother forms the full association between the object at which the 
mother warns and the reaction. Again, in a turkey, the I.R.M. of the 
bird of prey is much more complicated : form plays a role in it! If you 
make a dummy that has a short process on one side and a long 
process on the other side, and you draw the dummy across the sky on 
a string with the short process pointing forward, then the reaction 
is the reaction to the hawk, and if you point the long process forward, 
then it is a goose, and there is no reaction because a goose is not 
dangerous. Kratzig (1940) showed that grouse have two diff"erent 
reactions to two different birds of prey. They reacted to an eagle 
dummy by taking to wing and towering, and they reacted to a falcon 
dummy by taking cover. Kratzig has been reproved for doing too few 
experiments, but the whole point is that you cannot do many 
experiments because you get a conditioned reaction very quickly. 
When Tinbergen and I were experimenting on geese, one day, I 
came into the garden ; Tinbergen had prepared our dummies, and as 
I came into the garden he said, 'We aren't going to experiment 
today'. I said, 'Why not?' He said, 'I'll show you'. In order to attach 
our dummies to the piece of metal that slid on the wire it was 
necessary to climb the tree, and Tinbergen and I used to climb it by 
throwing our legs over a certain branch. Tinbergen went to quite 
another tree and threw his leg over its lowest branch — and all the 
geese looked up to the sky and went to cover. And that is why for 
I.R.M. experiments you ought to have a new animal for each trial. 


Dr. Lorenz, you mentioned three conditions under which diminu- 
tion of I.R.M. s might occur: hybridization, domestication, and ill- 
health. Can you differentiate among them in the sorts of damage done ? 
It seems to be conceivable that ill-health might diminish the strength 
of the response to each of a series of stimuH, and that you might get a 
quantitative but non-selective diminution, whereas in another case 
the diminution might be selective. I will give you an illustration of 
what I mean. In Bali I was in a village which turned out to be an 


area of acute hypothyroidism ; every single aspect of the culture had 
been simplified, but without loss of pattern. Instead of putting a 
hundred items in an offering they put twenty-five in, but they kept 
the essential elements — showing the difference between the diminu- 
tion that you get with hypothyroidism and that which you get with 


The question is very fascinating and I do not know whether I can 
answer it out of hand. There is in the reaction to the I.R.M. a loss of 
selectivity and that may result in making the release of the reaction 

A kind of vulgarization ? 


Exactly ! This would result in a person being sooner content with 
an object. That object need only possess a few of the releasing 
characters which act as key stimuli to the normal, undisintegrated 
I.R.M. , and a lot of others may simply be dropped out without 
impairing the object's releasing value. And you have no idea how very 
aptly the word you have just chosen, vulgarization, describes the 
situation. If you compare, for instance, the sexual behaviour of a wild 
greylag with that of a domestic goose in which the whole elaborate 
courtship ceremonies have become irrelevant, and in which any goose 
and any gander that are put together during the mating season will 
proceed to copulate without any more ado, you cannot help feehng 
that the behaviour of the barnyard goose is grossly vulgar. But I'd 
better tell you, in some more detail, the sequence of the disintegra- 
tion of an I.R.M. in domestic hens which Heinroth studied extensively. 

He found that the first key-stimuli which became independent 
were those of colouring. In the usual domestic hen that you find with 
peasants, a hybrid race of fowl, he found no hen that minded whether 
her chicks were black or white. In Phoenix hens, the long-tailed type, 
he found a good proportion of hens that would kill black chicks, and 
did not mind white chicks. In bankivas he found hens that would kill 
everything that was not wild colour. My friend Otto Koenig, in 
Vienna, tried to repeat those experiments with bankiva hens of rather 
doubtful origin to see if those things drop out in a few generations — 
we do not know how fast domestication occurs — and he did not find 
a single one that would kill a black chick, whereas there is no doubt 


that Heinroth's did kill black chicks. (A nice point is that they do not 
mind and are quite ready to mate with a black cock.) You find 
barnyard hens who will refuse to brood ducks and even kill the duck 
in the egg when they hear the first cheeps. Yet Orpingtons and Rhode 
Islands will take gosHngs and ducklings, so long as they are downy : 
they won't take a naked bird. If you let them hatch a raven or a 
cormorant, which are about the largest birds whose newly-hatched 
young are quite naked, they will kill them. Then there are Brahmas 
that will brood anything up to kittens. It's a nice sequence, isn't it, 
how the characters drop out singly. 


All these things Dr. Lorenz has been describing make one think 
that there are circumscribed mechanisms in the central nervous 
system which should be relatively easily susceptible to anatomical and 
physiological dehmitation. Has, in fact, any work of this sort been 
done ? Presumably there are centres where the energy for the I.R.M.s 
builds up. Where are they? 


I am not quite in agreement with the hypothesis that they must be 
circumscribed, for one reason: you must keep in mind that where 
reaction to form is concerned, as, for instance, in the turkey cock, the 
whole thing must somehow go through a Gestalt perception, an 
apparatus certainly very widely distributed inside the central nervous 
system. That does not apply to all I.R.M.s and, of course, there are 
all gradations including the case where one sensory organ only is 
required to elicit one response. If you have a cricket, where the female 
has a hearing apparatus with a range of only a few tones in pitch, 
and which is only meant to receive the mating call of the male (which 
releases a positive reaction causing it actually to jump into the loud- 
speaker in the celebrated experiment) there the question of selectivity 
does not arise; but when the jackdaw reacts in one way with a sub- 
missive attitude and in another way with an aggressive attitude, then 
the whole thing must be shunted somewhere over the remotest 
projections of the cortical visual apparatus. 


I was thinking more of the effector end of the mechanism. After 
the Gestalt has been received, brought in over certainly many path- 
ways, one would imagine that these then converge on some distinct 
anatomical structure which would be the power structure of the 



Some results of Professor Hess tempt one to think that in his 
excitation experiments he gets a place at least very near to the input 
on the receptor side (Hess and Brugger, 1943). In most of Professor 
Hess's experiments he got an entirely coherent, integrated sort of 
behaviour. If he stimulated the 'fighting centre', the cat behaved 
entirely as if there really was a rival. In such cases one would suspect 
that the excitation influenced a centre very high up in the hierarchy of 
the central nervous system, or even that it was the afferent side of the 
response that was stimulated. Such a cat behaves exactly as if it had an 
hallucination of another cat fighting it. But the interesting point is 
that Hess could get different levels of the hierarchic system of this 
instinct at will. When he let a very slight stimulation impinge at the 
more cranial point, he got a slight threshold-lowering of fighting- 
reactions. The cat would not yet attack in vacuo, but it would bite 
Hess's assistant whom it would not have bitten without that stimula- 
tion. In other words, slight stimulation at that more central point 
would put the cat into 'fighting mood' ; if the stimulus was increased 
the cat would attack substitute objects which were still less similar to 
a rival cat ; and so on, up to 'explosive' fighting in vacuo. But if Hess 
stimulated at a point situated a few millimetres farther to the caudal 
end of the brain, the cat, as a whole, was not put into a fighting mood. 
It got no threshold-lowering, went on purring peacefully, allowed 
itself to be scratched by the assistant, and then, on stronger stimula- 
tion, it would quite suddenly discharge disjointed fighting movement, 
like scratching and spitting, only to go on purring in the next moment. 
All this fits in with Tinbergen's theory of the hierarchic organization 
of instinct so beautifully that it is almost too good to be true (Tin- 
bergen, 1951). If you ask me where to look for the localization of an 
I.R.M., I should say: the I.R.M. joins on to the effector side of the 
response in those places where Professor Hess got his most generahzed 
instinctive behaviour patterns. 


We need, I think, to know whether there are any characteristic 
differences between mammals and birds in this field of ethology. We 
talked of loss of I.R.M. s. I want to know if there is any way of putting 
them back. That is to say, has anybody created in animals new 
I.R.M. s that didn't exist before? 


Birds and fishes were the first objects in which we found these 
things. A systematic study of mammals and insects in this regard is in 


its initial stages, but these initial stages allow us to state that it is 
quite surprising how these elementary mechanisms repeat themselves 
in animals as widely different as a rodent, a bird, a fish, an insect, and 
a cephalopod. Thorpe has found imprinting in insects which is 
absolutely comparable to that to be found in birds; the only thing 
about which we are still very doubtful is imprinting in mammals. 
Most mammals, except monkeys and man, are animals with a large 
olfactory region in their brain, and in these macrosmatic animals it is 
difficult to experiment because you can't control your stimuli — you 
don't smell yourself. There is the olfactory type of mammal and the 
visual type, and that is why the ethology of monkeys and of man is so 
surprisingly convergent with what we find in fishes, and very unlike 
that of olfactory animals. Nevertheless, we get more similar reactions 
than we expected, with the exception of imprinting. There are some 
instances where sexual activity in ruminating animals is optically 
released, and in these few cases it seems that true imprinting also 
occurs. All known observations concern sheep and cattle, but I am 
sorry to say none of them is very conclusive. 


I would like to close, if I might, with an anecdote which perhaps 
will relate, in a light way, the bird to the child. This observation was 
made by my son when he was about fourteen, on some cardinal birds. 
He had a bird-station just at the right photographic distance from 
our back window, and a young cardinal came to the station and began 
to eat the seed quite vigorously. Then his father flew down to the 
bird-station, and immediately this httle bird crouched, dropped his 
crest, lowered all his feathers, put his head up in a begging reaction, 
and would not eat anything on the bird-station unless fed by the 
father. When the father flew away, he roused, became himself 
again, stood up straight. This was repeated several times, and my 
wife and I also saw it. What I wonder is whether we can say that this 
was a regression to a dependent attitude in the presence of the father. 


Well, I don't think so. I rather think it is just the overruHng of the 
still stronger feeding reaction over the weaker one. I can duplicate 
that observation by one on young shrikes, exactly the same individuals 
on which I did the impaling experiments. These shrikes were fully 
fledged but were still fed mainly by begging. When there is anyone in 
the room the shrikes are prone to go on begging for a very long time, 
even when they are already quite able to eat. Now I went for a 
motorcycle tour for four days, and during these four days those young 


shrikes were left alone by themselves in my room. They fed themselves 
perfectly and they were absolutely healthy, sleek, and fat when I came 
back. I had important work to do, and sat at my desk and the shrikes 
were sitting in their cage begging at me, and I said, 'Confound you, 
you have shown me that you can eat for four days, I am not going to 
feed you any more'. In the afternoon I saw that the shrikes became 
seedy and sad and saw that they hadn't eaten one bit, because their 
begging reaction actually prevented them from feeding themselves, 
and they would have died of hunger because I was sitting in the 
room, though they would thrive perfectly when I wasn't. I think 
that's the chief explanation, and a more economical one, of the 



Electroencephalographic Development 
of Children 


When a child is born it exchanges a physiological for a social environ- 
ment. It is about some of the aspects of the child coupled with the 
social machinery that I want to talk. 

I am going to leave out of my remarks nearly all matters of tech- 
nology. For the most part these can be found in the hterature 
(Walter, in Hill and Parr, 1950; Walter, 1953). I am also going to 
take for granted that the information, such as it is, on adult neuro- 
physiology is accessible, though it may not be known to us all, and I 
am going to assert where I could, in fact, prove. You will have to take 
my word for much of what I shall have to say, and possibly in the 
discussions I can amplify some of my statements. 

First of all about the general difficulties of studying children. My 
wife and I have spent some years examining children from many 
points of view — electrophysiological, psychological, ethological — 
and we have constantly come up against the difficulty of assessing the 
nature of the population we are studying. Is it, or is it not, a select 
population? If one is going to study normal children of school or pre- 
school age, these children are necessarily drawn from schools, 
famihes, and friends who are willing for them to be studied; and that 
eliminates children whose parents are not willing — parents who have, 
perhaps, some quite natural superstitious distrust of science and 
scientists, and particularly of neurological institutes and mental 
hospitals. This eliminates a certain group of the population so that, 
when I tell you that this is a normal child or these are normal children, 
you must weight what I say with that previous knowledge. These are 
children with no neurological or psychiatric complaints, but they are 
inevitably selected for their wiUingness or their parents' willingness 
for them to be studied. 

The next difficulty which we have in considering children is that 


development of the individual physiology and psychology is coupled 
reciprocally with social influences. For example, many delinquent 
children come from unhappy and broken homes, but if you inspect 
those homes in detail you find that they also contain genetically 
pathological specimens. There are brothers or sisters who are 
mentally affected or insane, there is a father who is alcoholic, a 
mother who is a prostitute. The separation of social family factors 
from genetical and ontogenetic ones at the present time I believe to 
be impossible. 

Then, a question of pure technology which I must mention. I hope 
that many of you who are not physiologists or electroencephalo- 
graphers will be tempted to read encephalographic and electro- 
physiological literature, but I do beg you in reading it to be extremely 
sceptical of the results, particularly from the technical standpoint ; to 
bear in mind that the sort of information we collect from our 
machines is very liable to deceive us, that only the most refined, 
sophisticated and flexible techniques of recording and interpreting 
are hkely to be of value. If one goes through the literature on child 
electroencephalography, one finds very little that, in the last analysis, 
one can accept as fact. The methods of recording, the methods of 
display, the methods of transformation, of analysis, of statistical 
checks — all these are subject to very serious criticism, so that the 
assertions I am going to make about infantile and juvenile physiology 
are based mainly on my own work, merely because I do know at least 
exactly what the limits of accuracy and interpretation of that are. 
Beyond that, I would not like to go. The accuracy of even the simplest 
experiments, as you will see in a moment, is very limited. 

As you examine younger and younger children, the first feature 
which you find is a gradual increase in the general amplitude and 
profusion of electrical activity, and at the same time a progressive 
decrease in its frequency. The electrical waves from the head, recorded 
in the conventional manner, get larger and larger and slower and 
slower as you go down the age-scale. If you put electrodes on the 
belly of a pregnant woman, you can record before birth the electrical 
activities of the foetal brain, and these bear out the extrapolated child 
data : the brain activity of the foetus is, on the whole, very slow, 
irregular, and poorly synchronized. It can only be observed in 
snatches, because the foetus is always moving, so that there will be a 
few seconds when you get nice clean records, then suddenly there will 
be a sort of convulsion, the child will turn over and the head will 
have gone. 

If we follow our child through its first months of life, we find the 
slow activity, swelhng, waxing, waning, and in some stages of sleep 
and repose giving place to certain features well known in the adult, 


the spindles, 14 c/s oscillations characteristic of adult sleep. One then 
begins to notice what to me is one of the most fascinating features in 
the whole of this work, the enormous differences between individuals 
still in the normal range, a range so wide that I myself hesitate to be 
in any way dogmatic about what the normal E.E.G. is. If one looks 
for the meaning of these slow rhythms in children one sees that they 
are associated with the search for peace, if you like to call it that — 
the repose and inactivity of the child. As the child begins to get 
older, spends longer periods awake, pays more attention to its sur- 
roundings, makes apparently volitional movements and gives overt 
expression to affective states, one observes a decline in the slow 
rhythms. Often one hemisphere will start first to show suppression 
of this activity — sometimes it is the left, sometimes the right. At a 
very early age, sometimes as young as two or three months, long 
periods occur during which the slow activity is minimal, particularly 
when the child is attentive and being played with by its mother or 
performing some relatively complicated act. 

Now I am going to follow the various E.E.G. components up 
through the years and discuss what they seem to mean — supposing 
the components can be considered as individual phenomena. (It is 
quite possible, of course, that we are deaUng with a number of 
phenomena which have only a superficial resemblance.) Imagine us 
taking records longitudinally in children and following the develop- 
ment of the slow activity through to adult life. As the child gets older 
these slow rhythms — the delta activity — become more intermittent 
and in general they decline to a small figure of scarcely perceptible 
size some time between the second and tenth year. The situation is 
that illustrated in Fig. 15. At the top of the figure we have a chart 
showing the prominence of the various components of the E.E.G. — I 
use the word 'prominence' advisedly: it is a measure of both amph- 
tude and abundance of activity together as a product. The data have 
been obtained by automatic analysis of many hundreds of records ; 
the lines at the top and bottom of the delta band show the range 
of individual variations. If one plots from 100 to 75 as being the 
birth range, the variation becomes so wide at the age of about three 
that you can have one child showing almost no delta activity, and 
another showing as much as a newborn. 

Exceptionally this delta activity can persist until the age of eighteen 
or twenty. We have made a special study of this type of activity in 
delinquent children compared with normal children (Hodge, 
Walter and Walter, 1953), and we have been bold enough to 
suggest a term to describe the psychological features which statisti- 
cally and experimentally are associated with the presence in the brain 
of these diffuse, rather poorly synchronized, slow rhythms. This term 



is ductility, which means, of course, the abihty to be drawn out with- 
out breaking, for the personahty to be deformed and moulded into 
shape without cracking or taking a permanent set. Ductihty was 
observed first in deUnquent children, where we found the presence of 
slow activity associated with something that shocked me as a 
physiologist. It was not associated with childishness in the general 
sense, or with stupidity or intelligence, but with a relatively good 
attitude as judged by psychiatrists and schoolmasters — a good atti- 
tude to the children's mother, to their fellow schoolchildren and to 
their leisure. This was discovered by statistical analysis. 

As a physiologist I found this notion of a physiological basis for 
affection and spare-time occupations almost too good to be true. I 
was brought up in a very sceptical school of physiology where the 
study of the brain was considered rather disreputable, and normal 
psychological functions as completely outside the field of physio- 
logical investigation. But here one has a very firm statistical associa- 
tion between a measurable, quantitative, physiological phenomenon 
and a very complex and a very sentimental aspect of the child's life: 
the relation to his mother, to his spare time, and to his fellows. If the 
attitude were bad, if he were a naughty boy or a backward boy, then 
one would have accepted the relation as a phenomenon of immaturity 
in the nervous system, or inefficiency of brain metabolism, or what you 
like. But these slow rhythm boys were the boys who had a good 
attitude, they were the nice boys, docile, manageable, easily-led; in 
other words, ductile children. In the dehnquent population they were 
the sheep who followed the leader. We have tragic evidence that 
children with this character very easily find themselves inextricably 
stuck once they get into a criminal society. They may be pushed in any 
direction, and naturally tend to drift down the social gradients. The 
consequences of certain mixtures of dilferent types of this sort can be 
disastrous. Perhaps one of the most useful things we can do is to see 
how this work can be applied to such serious social problems as 
crime and delinquency. 

Slow electrical activity is common in adults in sleep, and I should 
like to put to you the hypothesis that the presence of slow electrical 
activity of diffuse nature (not associated, of course, with organic dis- 
turbance) is the external objective representation of mechanisms in 
the brain which are directed towards defence of the brain. We all 
know that in the brain the three mechanisms which in the rest of the 
body defend against the consequences of injury are totally lacking. If 
you break your arm you feel pain ; as a result of it the arm is immobi- 
lized and, even without the help of the surgeon, a better chance is 
obtained of the bone healing. In the brain you feel no pain. Secondly, 
you have no lymphatic drainage in the brain to drain away infected 


organisms or damaged cells. Thirdly, there is no regeneration of 
tissue; the brain-ceil once destroyed is never replaced. So the brain 
is in a vulnerable condition. It is enclosed in a hard box, but once that 
box is penetrated, or something happens in the brain system which is 
dangerous or undesirable, there are apparently no defences. I would 
suggest that the slow rhythms represent the one defence. The brain 
has either to shut off from excessive action a part of itself that is 
damaged or, in a normal person, to rock to rest those mechanisms 
of the central nervous function which are either exhausted or have 
attained a wrong set in the adaptive process. One can show that any 
comphcated mechanism which has its design not predetermined, but 
which determines for itself its own design, must have built into it 
quite elaborate mechanisms for what we call 'failure-to-safety' : that 
is, arrangements which ensure that should something go wrong, 
should some undesirable state occur, the mechanism shuts down as a 
whole and does not go too far wrong. I suggest that these slow 
rhythms are a sign of this functioning, as a 'failure-to-safety', of some 
of the brain mechanisms. 

The methods of studying these slow rhythms, particularly in 
children, from say two to twelve years old, are very important 
technically because the waves can be evoked and suppressed. Fig. 16 
shows two records taken from children both aged twelve. The upper 
one is an example of a record from an entirely normal child. The 
presence of the slow activity shows in the record, and the analysis 
peaks indicate the total quantity of slow activity during each period 
of ten seconds. The lower record is from another perfectly healthy, 
happy, normal child, but he displays in the periods of rest large 
regular slow rhythms at about 3 c/s, as well as an alpha rhythm at 
9 c/s. Whenever that child is stimulated by anything whatever, by 
sound, by a word, or even by an idea, the slow activity is suppressed. 
During this particular recording a flickering lighf was intermittently 
turned on, as shown in the bottom trace. Whenever the child was 
stimulated in this way, the slow activity ceased. This is a particularly 
old child to show such very rhythmic slow activity, and he was in 
fact very highly ductile. He was a charming child; any suggestion 
you made he would fall in with, any advice you gave him would be 
taken. His intelligence was quite adequate to take care of any normal 
problems, but he was easily led, and easily led astray. 

The next feature of children's brain rhythms which seems to be 
particularly important is the appearance and significance of the theta 
rhythms. These rhythms have a different frequency and a different 
time-scale. They are sometimes easily seen in the record, but some- 
times they are variable and subtle, and apt to be mixed up with other 
rhythms. In Fig. 15 the large band marked theta represents the 


FIG. 16 
__ yjy—— ^y^j — 




The upper record shows a characteristic but rather juvenile medley of rhythms in all 

areas. The lower is an example of an unusually large and responsive delta rhythm at 

3 c/s suppressed during the exposures to flicker. 

This type of record is associated with a high 'ductility' 

FIG. 17 



Sexes combined 
Foster and Anderson 


From Goodenoogh (1931), for comparison with graph 

of »he thaiva rythm development (Fig. 15) 
_J I I I •: 



Age in years 

growth-range of the theta rhythms in a normal juvenile population. 
The range is again extremely wide. Fig. 17 is a graph of the fre- 
quency of temper outbursts with age in children, taken from the 
work of GooDENOUGH (1931). There is an astonishing correspondence 
between the frequency with which temper is lost in children, and the 
appearance of theta activity. The period of maximum rate of temper 
outburst coincides not with the maximum height, statistically, of the 
theta activity, but with the maximum rate of change of theta activity. 
This period of most likely temper outburst is a period when the theta 
rhythm is changing most and is the period when it is first taking 
control of the organism, when it first becomes the most prominent 
rhythm of all. We have observed this experimentally as well, quite 
independently. One finds, in fact, that if a child of two or three is 
happy and content then the record may have little or no theta activity 
even with the most refined methods of analysis. The moment the child 
gets annoyed — and, of course, a child can be annoyed by an enor- 
mous variety of situations or stimuh which to adults are quite 
neutral — then one sees a burst of theta activity. This burst may long 
outlast the stimulus period. It may be a matter of hours, even in some 
cases of days, before the effect of a certain disagreeable stimulus has 


worn off. The first experimental situation in wliich we observed this 
was the following : we were taking records from normal children and 
in the younger children of two or three we found it quite difficult to 
get them to co-operate well enough for our purposes. We used to give 
them a sweet on a stick to suck, but for the purposes of the recording 
itself we had to remove the sweet in order not to confuse the electrical 
activity of the jaw-muscles with that of the brain. The moment the 
sweet was removed there was always this burst of theta activity. In 
the adult we found it more easy to evoke a theta rhythm by with- 
drawal of a pleasant stimulus than by annoyance. Real deep-seated 
annoyance is an almost impossible thing to contrive in the laboratory, 
but you can give somebody a pleasant stimulus and interrupt it, and 
then in the adult you get a burst of this childish frustration rhythm. 

I suggested that the slow delta rhythms represent a search for 
peace and the equilibration of the organism: I would hke to suggest 
that these theta rhythms represent a search for pleasure as a specific 
entity. This should, I feel, link up in some way with a psychiatric 
and psychoanalytic approach to pleasure and pain. 

I should like now to discuss the appearance and meaning of the 
alpha rhythms, shown also in Fig. 15. The distribution of the adult 
alpha types is first seen at about age nine to eleven; in other words, 
it is about this age that the distribution of alpha types becomes 
similar to that seen in the adult. I have to describe the situation in 
this somewhat roundabout way because of the individual variations. 
There are people who show no alpha rhythms at all, and there are 
those in whom the alpha rhythm is persistent throughout almost all 
their waking life whether their eyes are open or shut, whether they 
are reading, or anything else. Apparently the type of alpha rhythm 
relates to the type and vividness and persistence of imagery. If the 
delta rhythms represent a search for 'peace', and the theta rhythms a 
search for 'pleasure', I suggest that the alpha rhythms represent a 
search for 'pattern'. There are certain people who, when they close 
their eyes and relax, are not at rest. The moment the eyes are shut, 
a picture-show starts. These pictures — I speak from experience — are 
not in any sense obsessional, but they are quite vivid and can be 
turned on or off and manipulated as one chooses. In other people 
this does not happen. The time at which a juvenile population starts 
to divide into these types is about the age of nine or ten, which 
coincides with the time when reading ability increases considerably. 

The curves in Fig. 15 are simple statistical displays, garnered from 
a large population and shaken down by statistical methods into a 
fairly smooth curve; but if you follow an individual child through 
these phases you get an entirely diff'erent picture. Then there is 
nothing like a smooth curve. For some months the delta activity may 


go on showing occasional spindles ; then quite suddenly one day you 
get a burst of theta rhythm as the child wants something but does not 
cry; the first time the child is deprived, annoyed, frustrated, and does 
not burst into tears. 

The individual records show abrupt changes and also violent 
vacillations from one world to another, just like a switch which is 
hovering between contacts and, just while it is closing, chatters for a 
few days or hours and then shuts irrevocably. This is an observation 
which, if confirmed and worked out in detail, is very important for 
general ideas about living animals. The notion of the occurrence in the 
nervous system of the possibility of abrupt change helps us not only 
to understand, but to predict, the behaviour of organisms in a way 
which no other process possibly can. It means that you have a machine 
which can work within a perfectly well-defined framework of rules 
and which, if the representative point of that machine gets beyond a 
certain area, does not break down. There is suddenly a tick-tack-tock 
and a different set of rules is put up. The brain has a capacity for 
resetting itself, for setting up its own wiring, which, of course, con- 
founds the Cartesian dualists, who admit that you can make a 
machine to imitate any property of the human being, but not a 
machine to imitate the human being itself. I don't say whether you 
can or whether you can't, but what I do say is that you would have to 
make a machine that would reset its own contacts according to 
experience and the chances of survival. Many such machines would 
fail, but the machines that did survive would be self-adjusting 
machines, machines which could re-orientate their internal connexions 
according to what had been found satisfactory on a statistical basis. 
Such machines would not be logical computers, but computers of 
similarities and differences, of relations of patterns of environment 
and behaviour, readjusting themselves according to what had been 
found to work. 

Fig. 18a is a record which demonstrates some of the features I have 
described. It is from a girl of sixteen who was taking part in a test 
situation which involved her feeling, blindfolded, with her left fore- 
finger the outline of a groove in a block of plaster, which was in the 
form of a large letter H. With her right hand she was supposed to 
draw just what she felt, whether she recognized the design or not. 
On the top line of the record is her speech, her pulse, and her 
breathing : on the second line the resistance of her skin, the activity 
of her neck muscles, and the pressure of her pulse : on the third line 
the E.E.G. from the right and left occipital regions and then the 
position of the right and left hands. At the point H the H-block was 
inserted and she immediately began to feel and explore the groove, 
and began, rather badly, to draw the outhne with her right hand. The 


FIG. 18A 












The first record covers the start of the task and shows rather irregular breathing and 
complete suppression of alpha activity. The second record is a continuation of the 
first and shows a sudden sigh, a grimace, a violent drop in skin resistance, acceleration 
of the pulse, tensing of the neck and then a sudden increase in theta activity, all 
associated with a feeling of annoyance and frustration. 

FIG. 18B 

Showing the persistence and final subsidence of the theta activity as breathing, 

skin resistance and tension return to normal. 



Vw V..,-vw*^ kv*-^ V../^ V-,.w^^ Uv^iNMf^ 


4,5 T 10 




/^ yv^vv'"'''*"*^^ 

^ <1 ^VffcV»»^l^|;|<>i| V l^ * >*^'«»*''>Mi'i ' 

E.E.G. shows no dominant rhythm — her whole mind is active, 
working out this problem — then a little later she gave a great sigh, 
and grimaced, and later still, her muscles tensed, and the theta 
rhythm appeared, in that order. First came the breathing, then the 
grimace, the tensing of neck muscles, a drop in the skin resistance, 
acceleration of pulse, and, finally, the theta rhythm. At this point 
she made the remark, 'How far do I go?' She was worried about 
what she was supposed to do : the instructions given were quite vague; 
she had felt accurately the upright of the H and the cross-bar, but 
she hadn't found the other hne. Then she got really annoyed that she 
hadn't got good instructions — she was a schoolgirl and used to being 
told exactly what to do. 

Fig. 18b shows the further development of her theta activity. These 
are records taken continuously and you see that after the affective 
display you get a big peak which is the amount of activity at 5 c/s 
during than ten-second period. During that time the girl felt really 
frustrated, and didn't know what to do; as you see, her left hand is 
moving up and down, repetitively, constantly searching, and the 
right hand is not moving very much. The theta rhythm gradually, 
after a time, got smaller and smaller, and finally died away. Then 
there was another change in skin resistance and she returned to her 
normal state. Because of this frustration, she was quite incapable of 
solving this problem, one which every child that we have seen so far 
has been able to do. Her intelligence was quite high enough, but she 
got herself into a frustrating situation, and even at this age of sixteen 
developed an entirely infantile response of theta rhythm which was 
the essence of frustration for her, and which was the signal not to 
finish or solve the problem, but to get out of it, to regress to an 
emotional display. Then after the affective discharge, skin-resistance 
changes, the theta rhythm, the desire to get out of it all, you see her 
suppressing this display and making up her mind to be a good 
girl. She was a good schoolgirl from a good school, and we were nice 
people and she was not going to lose her temper on this silly problem. 
After she had got over this she solved other problems quite well. It is 
very characteristic during testing, as you knov/, in children of this 
age, that they may make several silly mistakes in the ordinary intelli- 
gence test, and then later on solve problems up to their age level 
perfectly well : this is just that kind of emotional instability that one 
sees at that age. 

Fig. 19 has to do with another aspect of personality; the extent to 
which the alpha rhythms vary from time to time in the individual. 
This suggests the notion of versatility as a personality parameter 
which is at least measurable. Some brains keep the same tenor of 
activity hour after hour, even from minute to minute, from second to 


FIG. 19 






FRIEND - PRY - - - 




Above. The frequency analysis averaged over forty seconds during rest. There was 
little variation from epoch to epoch in the profile of the spectrum. Below. The 
conditions during the word-association chain test. The range of ideas was limited 
and each link was associated with a fall in skin resistance. The E.E.G. analysis shows a 
slight shift to the right in the modal peak but the profile as a whole is little altered. 


second. They have a small repertoire; little versatility, The other sort 
of brain may not be any better, in the sense of having a higher IQ, 
but is one which is constantly changing. This is something we can 
measure quantitatively and automatically and objectively. The record 
of Fig. 19 is an example. In it a girl, under a test situation, has her 
speech recorded at the top and her skin resistance below. The group 
of upright lines is the frequency analysis averaged over the previous 
forty seconds of activity. You see the profile of one spectrum is very 
much like the profile of all the others: this was a child with low 
versatility. The lower record shows the effect upon the same girl of 
the task of forming a word association chain beginning with the word 
egg. The record was taken toward the end of her association. At this 
point she had got to 'friend', 'pry', 'creature', 'caterpillar', 'insect'. 
The E.E.G. analysis shows a very similar profile to the resting one, 
except that the mode of the analysis, the highest point, is 11 c/s 
instead of 10 c/s. You may notice also that for each of the association 
changes there is a large drop in skin resistance. At the word 'cater- 
pillar' she has managed to get out of the affective trap of 'friend, pry, 
creature'. You can see a slight tendency to produce theta activity 
which is suppressed, because she got on to the notion of caterpillar, 
and the chain was terminated just there. 

Investigations of this sort on children have so far given fairly 
unequivocal results, showing that versatihty can be measured from 
a very early age and, as far as we have seen, a child that is versatile 
in respect of one sort of rhythm will also be versatile in other respects, 
suggesting that this is a characteristic of the nervous system which 
can be measured and extrapolated with some safety. Versatility may 
have some connexion with the sort of things people like to do and 
particularly with the way people join up in groups. Two versatile 
people don't generally get on very well together; there is too much 
criss-crossing of ideas, the association is explosive and unstable. A 
versatile person and a non-versatile person get on very well together, 
because the versatile one will feed the other with ideas. The electro- 
physiology of these social groups seems to me one of the most 
fascinating things that one can study. 

I have put before you these notions of ductility, of temper-keeping, 
and of versatility as personahty factors. Now I am going to consider 
another factor which I suppose one could call stability or balance. 
This notion is related to the Pavlovian analysis of behaviour. Pavlov 
identified a character which he called balance in his animals and in 
his human subjects. This is important also in brain physiology, 
because one can show that the way in which an organism responds 
to violent stress varies all the way from a 'freezing reaction' of doing 
nothing at all, to a violent and sustained oscillation. 


FIG. 20 





This discharge was accompanied by a transient lapse of awareness. This type of 
record is associated with considerable 'instabihty' and is commonest in epileptics 

In Fig. 20 you see the record of a neurologically normal child who 
was a delinquent. He was at this time sixteen, and two years later he 
was convicted of a capital offence. He was one of the ductile type 
and he was also unstable. He had run away from home and required 
firm institutional control. This record is the response to a series of 
flickering light stimuli, which produce a gradually increasing fast 
discharge and then a large slow discharge. The whole brain is excited 
into an almost convulsive state by this simple flickering Hght. This is 
a response which one can elicit from about one in thirty of the normal 
population and is associated with the type of behaviour one sees in 
very young children. It is inappropriate, exaggerated behaviour. It 
is found in the sort of person who commits not a motiveless crime, 
but who instead of shouting at somebody or insulting them, knocks 
them down. If you get a combination of a lad with high versatility 
and normal E.E.G., and a boy like this with a high ductility and high 
instabihty, then almost always crime results, often murder. 

In Fig, 21 we have a similar sort of picture. This is a record of a 
normal girl aged twenty, but here the stress has been made very 
difficult to bear for anybody, by making the stimulus itself depend 
upon the brain activity. The brain activity is made to operate a 
trigger circuit which provides the stimulus to reaff"ect the brain 
activity, and a feedback oscillatory system can be set up. Most brains 
are equipped to deal with this, which very likely occurs in the normal 
course of events. I suggest that it is one of the functions of the slow 
activity of the brain to deal with precisely this situation. Here we see 
the activity in the left occipital region arranged to provide the 
stimulus which is shown on the bottom trace. This bottom trace 


FIG. 21 


but less extensive and protracted, evoked in a normal adult by feedback 
flicker, the left occipital rhythms being made to generate a flash-stimulus. 






The large slow wave (underlined) interrupts the reflexive mechanism, but the 

subject complained of feeling iight-headed' whenever it occurred. The effect 

faded after about 15 repetitions. 

copies the occipital one, and adds to it the stimulus, the sharp upward 
deflections indicating flashes of light which produce in this normal 
girl a violent discharge over the whole brain, leading up suddenly 
at the marked point to a series of very sharp discharges. Then these 
are cancelled in a large slow surge. The feedback chain is broken and 
this is what I mean by failure-to-safety. This may result in uncon- 
sciousness, and most people in this situation feel a bit 'swimmy'. 
This was the first time that this person had had this experience. If 
you repeat it, the effect no longer happens in a normal person. About 
a dozen experiences of this sort produces a gradual fading of both 
the fast and the slow response ; in other words, the brain adapts itself 
in such a way that the retroactive networks are no longer established. 


I believe failure-to-safety to be an important aspect of pet it-maL 
The reason why the wave and spike of petit-mal persists and recurs 
is that the brain does not reset its switches so as to avoid the retro- 
active arrangement. The mechanisms one sees at work here, I think, 
are those responsible in the brain for the four selective operations 
which are essential for the preliminary stages of learning. I think that 
these basic processes originate in the reticular system of the brain 
and are connected particularly with the estimation by the central 
nervous system of to what extent a set of signals are significantly 
associated. We are slowly collecting evidence for the view that the 
various operations necessary for learning develop at different rates 
in different children at different times; one may get, for example, a 
very mature and high selectivity in a child with no constructivity at all. 
The child has a very good discrimination and yet cannot retain in its 
mind a notion that two things are related. Conversely, a child may 
have a very highly developed constructive ability and yet be com- 
pletely unselective. 


You said that there was a certain correspondence in electro- 
encephalography with our stages, but I must admit that I have not 
quite grasped where this correspondence lies and should like to have 
some details. 


It is very difficult to describe briefly what this correspondence 
is, and our own experiments are very far from complete. In stage I (of 
early infancy) wherein Mile Inhelder distinguishes six different 
epochs, transitions from the second to third epoch, from the primary 
circular to the secondary circular reaction, seem to be related to the 
way in which the slow delta activity is interspersed with faster 
activity. The change from epoch three to four, from the secondary 
circular reaction stage to the co-ordination of patterns, where you 
get for the first time a real goal-seeking mechanism, seems to be 
associated with the time at which the electrical responses are not 
directly evoked sensory effects but appear in the temporal and frontal 
lobes as an abstraction of the stimulus. At this stage, instead of a 
response to flashes of light being merely in the visual area there will 
appear in the temporal or frontal lobes transient interwoven crystals 
of electrical energy which maintain for seconds or minutes the abstract 
form of the stimulus. The age at which this first happens is quite 
critical. Before this the signal reaches the sensory area, the child sees, 


but it doesn't know the meaning of what it sees ; then quite suddenly 
it does see the meaning. This is, of course, related to the abihty to 
read and understand. 

These experiments are rather tedious to do because they must be 
longitudinal; cross-sectional surveys are very misleading in this 
matter. If any of us are going to go on with this sort of work it's 
going to cost a great deal of money, because we have got to follow 
many children for years, examining them in enormous detail by 
methods that are quantitative, objective, and above all unobtrusive, 
because in making these experiments it is essential that one should 
not inject into the child the sort of thing one wants. It may be very 
difficult to work out a technique whereby you are not coupled too 
closely to your system, not yourself part of it, influencing it all 
the time. 


Don't you need really to develop children that think that the whole 
of this measuring situation is part of life? 


Yes, of course, just like going to school. 


I have a small point of criticism about your relating the curves of 
growth of keeping the temper from one set of data (Fig. 17), and 
electroencephalographic frequency from another (Fig. 15). I feel 
very dubious about relating two cross-sectional-type curves in that 
way. If the data are not taken on the same child followed longitudi- 
nally, the appearance of correlation may come about owing simply to 
general changes with age, though the correlation between the two 
variables in any given child may still be very small. 


That is a most reasonable criticism. In Figs. 15 and 17, I agree 
entirely that such relations can be quite meretricious. But we had 
already established experimentally a relation between theta activity 
and bad temper; the similarity between the two curves struck me as a 
particularly happy coincidence since the data were culled from 
quite independent sources, while our experiments with teasing children 
may have been spoiled by some sort of unintentional collusion. 



I should like to know if Dr. Grey Walter can mention some sort of 
a pathological condition in a child which more permanently changes 
its electroencephalogram to a slow record and if this is accompanied 
by the related psychological manifestations. 

Another question: we have seen how different emotions can 
influence the E.E.G., for example, how an angry child will show a 
changed type of activity. I have myself been doing some work in this 
field and found out how anxiety can influence the records and increase 
the amount of theta and also delta waves in the record. My question 
is : Can a child exposed to constant anxiety have its electroencephalo- 
gram permanently changed ? 


I can't give an unequivocal answer: it is rather difficult to get 
children who have been through a traumatic psychological experience 
and arrange to observe them long enough and carefully enough to 
detect these changes and measure them; however, we have had the 
opportunity to see a few children as inpatients in whom just such 
changes have been observed. There was, for example, a boy of about 
twelve who had one of those pecuhar psychometric documentations 
in which the results of various tests seem to be incompatible. The 
most striking feature about him was his inabihty to read. He was a 
very intelligent boy by non-verbal tests, with an IQ of about 130. 
His E.E.G. showed terrific profusion of alpha rhythms of all types. 
He had absolutely no visual imagination; but he had been taught to 
read visually. At the same time as these alpha rhythms he had a lot 
of theta activity and a lot of delta activity; he had an E.E.G. which 
would be more or less normal for a child of three. During the fort- 
night he was admitted, the E.E.G. record was completely trans- 
formed ; his alpha rhythms became moderately responsive, theta and 
delta activity disappeared. His whole attitude to the problem of 
reading and the extraction of concepts from the mnemonic marks on 
paper was completely transformed by a process of non- visual tuition 
and separation from his family. Five years ago I should have said 
that records like this were unalterable. At one time I supported the 
statements of Lennox, in America, who said that the E.E.G. is an 
hereditary trait (Lennox, Gibbs and Gibbs, 1945), but experiments 
we have done since that time have considerably modified my ideas. I 
have seen in E.E.G. records and in the psychological state of a child 
both regression as a result of injury followed by recovery, and also 
transformation from infantile behaviour and an infantile record, to a 
completely adult appearance as a result of quite mild psycho- 
therapeutic treatment or the mere passage of time. 



I have been fascinated by Dr. Grey Walter's communication and 
should like to ask him two questions : 

(1) Does the troubled stage you speak of occur before six years: 
that is to say, does it precede the first appearance of the alpha 
rhythm or does it come at about twelve years when the alpha rhythm 
becomes more regular and stable? 

(2) Do you interpret the theta rhythm as indicative of stabilization 
of behaviour with attainment of control over self or as a phenomenon 
showing the adaptive reaction of a particularly emotive child? Dr. 
Monnier and I have examined a child of six years who showed a great 
deal of theta rhythm and we wondered what was the reason for this 
rhythm. Is it a question of a pronounced reaction to emotivity or of a 
stabilization of rhythm natural to this age? 


Was this spontaneous or deliberately evoked ? 


It was spontaneous, but very much increased by hyperventilation. 
Why do you consider now the amount of theta rhythm as an expres- 
sion of controlled temper rather than as the expression of affective 
bursts or increased emotionahty as you described it previously? 


This is a change in my own point of view, based on what I have 
seen. The theta activity is associated with the practical need to control 
temper ; in other words, in the child you speak of, you have a child 
who has a bad temper, and because of that, he has to control that 
temper, and it is the effort to control it which seems to be associated 
with theta activity. But there is something else underneath which we 
must study in great detail, and that is the possession of a bad temper, 
whether or not you can control it. 

In answer to Mile Inhelder's first question about the appearance 
of the alpha rhythms, it seems that rhythms of alpha frequency may 
appear well before six, in the back of the head ; responsiveness to 
visual activity starts considerably before six in some children. You 
can sometimes find signs of alpha rhythms even in babies of a few 
months, but at the phase at which it appears and becomes responsive, 
there occurs usually a considerable vacillation in the child's behaviour. 
He will suddenly be able to read frightfully well, and then the next 
day he can't read at all. Or one day he will play with his toys and 


bricks like a baby, and then quite suddenly he will go on to draw a 


This notion of ductility has puzzled me quite a bit since I first 
heard about it. What occurred to me was that the good child is not 
necessarily a normal child. Normal children are difficult, especially 
in certain situations, when they have their emotional and instinctual 
responses aroused in regard to special people, particularly parents. I 
wonder whether these ductile children haven't got some of their major 
social responses knocked out. One knows, of course, that that does 
happen to children in hospital. They are very difficult for a bit, then, 
after a few days or a week or two, they 'settle down' and are good. 
And everyone says, 'Isn't it nice, they are good'. They are being 
ductile, but they are also being very abnormal in the sense that, as 
people remark, 'They forget their mothers'. I was interested in the 
notion that the delta waves are related to a search for peace, because 
that is how one conceives of the separated young child : he gives up 
wanting his mother. To want and not get her is much too turbulent 
and awkward an affair so he 'forgets' her : he then becomes ductile — 
he has found a sort of peace. 


Dr. Sylvia Klimpfinger found in hospitalized children who were 
torn from their mothers, and who were exposed to a quick change of 
nurses, that they tried to form personal contacts with the new nurse, 
in a much weaker way the second time, in a still weaker way the third 
time, and then they gave it up. These children subsequently showed 
very difl'erent behaviour and could be divided into two types. One 
type was all over everybody, and the other type behaved hke a chow 
dog that has lost its master — they became autistic. All of them showed 
a very great instability in their interests. They couldn't be kept at the 
same play for any appreciable period. They behaved then, especially 
the autistic ones, very much like delinquent children, with a general 
weakness of emotional reactions. These children who had their early 
reaction-to-mother knocked out were so similar to children who, 
either genetically or partially genetically, had a weakness of emotional 
and social reactions, that you could not tell them apart. 


The point against the explanation of ductility that Dr. Bowlby 
suggests is that statistically the children who had the most prominent 


slow activity were those who had preserved most nearly intact a good 
attitude towards their mother: they wrote letters to her, they asked 
to see her, went back to stay with her during holidays, and so on, 
and during their leisure time they were the most socially active ; they 
made models, they played games, and they were good company to 
their fellows ; in fact their social orientation was most nearly normal. 
They were not those who rejected the mother-figure or had become 


Was this information based on an average of a questionnaire 
answer ? 


The isolation of this factor was based on the statistical results of 
an inventory of psychological and sociological inquiries. 


Perhaps one would need to have a group of the typical ones 
examined more intensively. I think you might find that their relation- 
ship to their mothers was not as appeared in the statistical results. 


Yes, that might be so. These studies were made about three years 
ago ; since then we have made a few more intensive studies and a few 
longitudinal studies, and as far as we can tell this still holds true: they 
are the children who have not rejected their mothers, who still keep 
in touch with them. One interesting thing we found was that the boys' 
attitude to their mothers was not necessarily associated with the 
mothers' attitude to the boys. The mother might be completely 
indifferent to the child, but the child would still be fond of the mother. 
That is not true of the father relationship. The relationship of the 
father to the child and the child to the father was reciprocal. If the 
father liked the boy, the boy liked the father. 


I think that the extraordinary differences in the behaviour of these 
people in different social situations is most perplexing. In one situa- 
tion they behave very satisfactorily and then, when they get into an- 
other situation, they go into quite a different gear. I am thinking of a 
well-known murderer who was head-boy of a delinquent school. He 


was obviously an exceedingly psychopathic boy, but in that particular 
environment he could function as a social being; whereas in another 
situation where other stimuli reached him, such as sexual stimuli and 
love-object stimuli, his behaviour was quite different — he murdered 
his girl-friends. 


I wonder whether this ductihty question isn't a matter of simplifica- 
tion of the response pattern. Perhaps the response of these children 
is not less normal, but less in quantity than that of a non-ductile 


What is the relation between this measure of ductility and measures 
of suggestibility such as the Eysenck sway test ? 


I don't think the psychologists used that particular test, but they 
did use a number of others and it looked as though certain aspects of 
suggestibility (the results, for example, of the Diiss test) would relate 
significantly to ductility. 


There are characters we call in jargon 'working idiots' in New 
Guinea and Melanesia. The only idiotic thing about them is that they 
work ; they are people who will do things just because you ask them 
to. They are not economically forced to do this. They are not 
stupid, and they are quite well recognized through quite a large 
cultural range. This is against Dr. Bowlby's explanation as com- 
pared to a genetic one, because they occur in societies of such 
extraordinarily different mother-child patterning. People regard 
them as a resource. My suspicion would be, from the distribution in 
New Guinea, that there is a genetical element. 


They might not be really parallel to ductile children. 


I have three questions. The first is, that I understand that the alpha 
rhythm has its main centre, as far as the cortex is concerned, in the 
occipital area : I was wondering if the theta was similarly associated 


with any particular part of the brain? The second question is: is 
there any relation between the type of rhythm and whether the 
subject tends to use auditory or visual imagery ? The third question is, 
in view of the difficulty of comparing unsophisticated people with 
sophisticated, and the difficulty of making the African calm and at ease 
in a test situation which is very frightening for him, would there be 
any advantage in comparing these people in sleep ? Does sleep vary 
during life in any constant pattern as one grows older ? 


The distribution and geometry of the alpha rhythms are extremely 
variable, but certainly they are oriented longitudinally over the back 
of the head rather than laterally at the side. There is a relationship 
between the distribution of the alpha rhythms (which are always 
plural) and the way the information coming into the brain is distri- 
buted and relayed around the surface of it. Now compare that to the 
theta geometry. There are many sorts of rhythms in the theta, but the 
most prominent one is at right angles to the alpha field; the alpha 
rhythm is fore-and-aft at the back, the theta rhythm athwart the 
head at the side, mainly in the temporal and parietal regions. In 
certain parts of the brain these two interlace. At a certain age of the 
child you may have alpha and theta activity going on at the same 
time. The part of the brain where we see sensory signals first 
abstracted, or given meaning, is the part where this interlacing is 
usually most intricate, where the texture is most elaborate. You get a 
sort of tartan of these spontaneous rhythms, and it is where this 
tartan is first formed that you get the most highly abstracted and 
elaborated and preserved pattern of incoming signals. In other words, 
it looks — on a completely superficial and possibly mistaken view — as 
though the pattern-seeking and pleasure-seeking mechanisms were 
interlaced in some way, like the threads in a sewing machine. 

In relation to the question of examining Africans or any people 
with a different background, this is obviously most important, be- 
cause, in order to compare two populations and get any idea at all 
of what an African's brain is like, one has to make an extremely 
elaborate analysis of the whole situation, to see how he looks at it. 
In other words, you must be an African or train an African to study 
his own people; it seems to me that is the only way we can do it. One 
simple example of that: most people in laboratories wear white 
coats; for most children a white coat is a danger-sign. The result is 
that if a child in our laboratory is examined by somebody in a white 
coat you may get one sort of a record. If he is examined by someone 
in a green coat you may get an entirely different record. 


To go on with my answer to Dr. Carothers, I don't know very 
much about rhythm and imagery in Africans. In an English popula- 
tion, about 70 per cent of the population have a mixed imagery 
type : they can turn on any type of imagery they need, more or less, 
within reason; about 15 per cent can only use visual imagery, they 
tend to be obsessed by visual images, and can't do much else; and 
about 15 per cent cannot produce a visual image if you pay them. 

The other question was about sleep. 


Sophisticated and unsophisticated people studied in sleep. The 
sleep pattern as it is described in textbooks is not much like the real 
sleep patterns, which are much less rigid. There is enormous personal 
variation in pattern, but there are certain features that do occur very 
commonly. Always at some stage you get slow rhythms, always you 
get spindles. The first stage is usually a theta stage, which is a 
'floating' stage when you feel your body disappears and you are 
floating — that is usually the time when the afferents are cut off from 
the body — but the alterations of these phases, and the amount of 
time occupied by each one during diurnal or nocturnal sleep, is an 
enormously personal factor. It seems to be set at a very early age and 
to maintain itself throughout life. You can predict to a certain extent 
from the examination of a child's sleep record the sort of sleep he 
will have when he grows up — apart from emotional disturbances and 
so on ; and even then you can predict, to a certain extent, from this 
very early infantile characteristic what effect such disturbances wiU 
have on a person's sleep. We don't know whether there is any ethnic 
variation in these characters, or if the situation is different in peoples 
who can sleep very easily. Nor do we know whether sleep patterns 
are susceptible to conditioning. We know from Pavlov that sleep 
can be conditioned ; it is one of the standard effects of delayed and 
trace reflexes. But even when tried out in dogs it is found to be 
characteristic of the individual dog, and to be one of Pavlov's 
typological criteria. 


But is there no close parallel between the degree of development 
during waking life of the individual and the rhythms in sleep during 
the same period of life ? 



You mean as a general rule throughout? No, you may get the 
sleep variations and the adult variations behaving quite differently. 
I don't say there aren't some correspondences, but superficially 
considered the two things are different mechanisms. That is probably 
simply because we haven't used the right sort of stimuli in waking life. 


To what extent do you believe that the different electroencephalo- 
graphic patterns could be related to Rorschach studies ? Just lately, 
in Yale, they have developed differential patterns of the Rorschach 
in different ages of childhood (Ames, 1952) and it struck me there 
might be a parallel, for instance, between form and percent alpha 
or maybe the Erlebnistypus and the delta pattern and so on. I feel 
that this might give us an interesting slant on our problem. 

grey Walter: 

We have data on a number of cases. There was a boy who had a 
number of fits whose Rorschach was grossly abnormal when first seen. 
His E.E.G. also was highly pathological, immature, and retarded, 
and showed some epileptic features as well. He was then given anti- 
convulsant drugs, which controlled the seizures quite well. Three 
months later he was seen, at the age of about eleven years and three 
months. The Rorschach was repeated, which, of course, was a doubt- 
ful thing to do, and there was complete transformation to normal. 
His E.E.G. also was then normal. We thought this might be due to the 
drugs, so we took him in as an in-patient, withdrew his drugs for two 
weeks, and though he did have one seizure, his whole personality 
remained quite stable. So in that case within the space of three 
months we were able to see a complete transformation from an 
immature, quite pathological, child with an abnormal Rorschach, 
abnormal behaviour and abnormal E.E.G., to an integrated, almost 
adult person. 


I should Hke to say a word on the question of choosing longitudinal 
or cross-sectional series. It is true that statistical population studies 
found in the hterature are mainly cross-sectional studies. Today it is 
considered that longitudinal studies, even when much more restricted, 
are of much greater interest. Personally I consider that certain cross- 
sectional studies would continue to offer great advantages where very 
homogeneous groups are used. 


At present I have the opportunity of examining a group of 200 
persons. They are cadet pilots of twenty to twenty-four years who 
have been medically examined and have the physical and intellectual 
aptitudes qualifying them to become pilots. They form, then, a very 
homogeneous group. We insisted on getting this group because the 
Air Ministry in Paris asked us to define the criteria of E.E.G. nor- 
mahty for acceptance or rejection of cadet pilots. For a long time we 
refused to take such a responsibility, but during a recent experiment, 
when we were studying the effect of stimulation on the electro- 
encephalogram in an attempt to find proofs for the existence of an 
epilepsy of which there was httle evidence, we noticed that the 
threshold above which these cadet pilots responded to stimulation 
seemed much lower than that which had seemed to us to be normal 
up to then. The study of about fifty individuals taken at random and 
from different age-groups had in fact given us a vague statistical idea 
of what was the normal threshold. 

These very low thresholds of stimulation seemed very curious and 
we wondered why these individuals were, in relation to our previous 
work, pre-epileptics. Our psychologist friends, having considered the 
problem, then suggested that our subjects had wanted to become 
fighter pilots because they had this particular characteristic which we 
considered as being pre-epileptic. 

Recently when we began to examine a new group of 200 cadet 
pilots we were surprised to discover again, even from the first ten, 
that the average electroencephalogram and the average reactions to 
various sensory stimulations were very different from what we had 
thought up to then to be the normal aspect for that age. We found 
with this group an abundance of slow waves in the parieto-occipital 
regions and a notable quantity of rapid rhythms in the anterior 
regions. These rather special characteristics make us say: take care, 
this is perhaps an abnormal individual. If we took his epileptogenic 
threshold we should no doubt find it very low; if we examine him 
thoroughly we may find traces of a past pathological attack, although 
the history has allowed us to think there has been none. 

Given the very homogeneous constitution of this group, which is a 
selected group, the common characteristics we shall find among these 
persons will be far more valuable and significant to us than if we had 
carried out this study on a cross-section of the population of the same 
age, but taken at random. 

If I might add another conclusion I would say that this cross- 
sectional study will cost less than a longitudinal study on say 10 per 
cent of the same subjects. This study will become more valuable 
when it is compared with other studies of very homogeneous groups 
such as we have already begun to consider. Thus we are going to be 


sent a group of French railway employees chosen to be engine 
drivers. Naturally, we always choose individuals without, or with very 
little, neurological history, and I think that we can thus obtain some 
guiding points for discussions, I will not say on the normality of an 
individual but on the aggregate of his special characteristics. 


I agree with Dr. Remond, but I think two things should be speci- 
fied : on the one hand, the necessity of constituting a vahd sample, 
and on the other, for a study in genesis, the need to proceed 

I do not think Dr. Remond's arguments give a full reply to the 
problem of choice between transverse cross-sections and the longi- 
tudinal method, since the example he has furnished concerns adults 
and the problem for him is to choose a significant group among these 
adults. I therefore think that the question remains open and that 
although in some cases the cross-sectional method may suffice, in 
others it may be very dangerous. Thus, if we study girls of thirteen 
the cross-sectional method is meaningless, because some will be 
pubescent and others not. In this case the cross-sectional method 
cannot be used. Therefore I think it is necessary to distinguish 
between the two questions, that of the constitution of the sample and 
that of the choice between the longitudinal method and the cross- 
section method. The question of sampling always comes in but it can 
be differently solved according to the method chosen. 


I should nevertheless like to say that the first series in which we 
were interested comes within the bounds of this group's study because 
the growth of the subjects is not finished. I think the end of growth 
comes at about twenty-five years. However it may be from other 
points of view, as regards electroencephalography this group was 
shown to be relatively infantile. We do not think that these charac- 
teristics are any more evident among younger children. Moreover. 
this group will help us define one of the particular aspects of the end 
of growth or development. 



Stages of Psychological Development 
of the Child 


I intend to comment briefly on the report drafted by Professor 
Wallon, thus coming back to the stages of early childhood, especially 
to the first year. After that I will go over the stages in the period of 
schooling; then I will take two sectors of behaviour, the evolution of 
graphic ability and the evolution of language, and refer very simply 
to the main stages, as they have been defined by a large number of 
authors. Then I will conclude with some considerations on method 
which seem to me common to psychology and all the branches of 

I have been asked to prepare a report on the work of the Wallon 
School and also on some of the work of Gesell, whose pupil I was. It 
seems to me already significant that Mademoiselle Inhelder and 
myself should have been asked to give an account of certain psycho- 
logies — I am purposely using the term in the plural : the psychologies 
of Wallon, Piaget, and Gesell, It seems that there is not a science of 
psychology, but a group of systems. We can, however, ascertain by 
reviewing psychological literature that there are extensive data on 
which to base an objective science. It is, therefore, not lack of 
estabhshed facts which leads us to speak of psychologies in the plural. 
Because of the nature of psychological facts, and perhaps certain 
biological facts too, organization and interpretation are constantly 
necessary and these established facts, if they are to be used, require a 
system in the present state of our knowledge. 

Motor Development in Early Childhood (according to Professor 
Henri Wallon, 1925, 1946, 1947, 1949) 
There are two ways of presenting the motor development of the 
child. The first is to give a description of the child's reactions in the 
chronological order of their appearance and in relation to their 
physiological context as well as to the circumstances of the moment. 

II 161 

This is the basic method ; but it calls for much detail and can lead to 
confusion owing to possible anticipations or frequent overlapping 
between manifestations differing in significance. The second is the 
functional method which groups these manifestations according to 
their nature. It is synthetic and more interpretative, but also more 
distinct and exphcit. It is the method adopted here. 

(1) Automatisms of Posture 

Co-ordinated reactions of the head, the limbs, and the trunk in 
response to certain exciting causes can be observed from the foetal 
period onwards. Such, for example, are the labyrinthine and cervical 
reflexes described by Magnus and de Kleyn. They are easily demon- 
strated in premature infants and also, under favourable conditions, 
in those born at term, by rapid displacement of the body upwards, 
downwards or horizontally, or by change in the orientation of the 
head relative to the trunk. They soon disappear. It would seem, 
nevertheless, that this series is connected with the reactions consisting 
in adjusting differently the segments of the body to each other and to 
exterior supports. To begin with there is the reaction acquired during 
the first few weeks : the active hfting of the head and its lateral dis- 
placement by means of the muscles of the nape of the neck. After that 
comes a whole series of readjustments arising from the child's 
contact with the ground in attempts to sit down, to stay on all fours, 
to stand up and finally walk. The difference between these reactions 
and the reflexes of Magnus and de Kleyn is that instead of being 
purely passive they are the result of active ventures into the outer 
world. It is the same difference as exists between intra-uterine Ufe 
and Hfe in free space. The automatic and unreasoning character of 
the first reflexes can be found again in the sudden emergencies 
of physical danger; they are components of actions with a fixed 

(2) Tonic Activity, Attitudes, and Means of Expression 

It is difficult to distinguish at birth between the spasms arising from 
tonic activity and actual movements. Cries, and the accompanying 
gestures — particularly stiffening of the trunk, opisthotonos, and 
sudden tensing and relaxing in the arms — appear to be definitely 
more spasmodic. The motor discharges in the lower limbs have the 
more dynamic character of pedaUing. Perhaps there is a period where 
tonic and clonic activity are still poorly differentiated; this could no 
doubt be discovered through electrical examination of the muscular 
apparatus. It is, however, tonic activity, both of the visceral muscles 
(respiratory, laryngeal, and gastro-intestinal) and of the skeletal 
muscles, that corresponds to the effects noted. Its proper field is that 


of attitudes and expression; in other words, that which is of greatest 
use to the child, since only by these means can he obtain and sohcit 
from his surroundings the help which is essential. The visible effects 
of his needs, his discomforts, and his sufferings soon hnk up, by 
means of conditioned-reflex mechanisms, with the tutelary effects 
which they have caused, and eventually become a means of arousing 
them. Towards six months the expressive manifestations of the child 
are so finely graded that all the major varieties of emotion can be 
distinguished. Vocal intonations precede language. The voice, the 
medium of speech, is a tonic activity. The attitudes which result 
essentially from tonic activity establish contact between the individual 
and the corresponding situation, underline the meaning of this 
situation and contain already its image. 

(3) Circular Activity 

There comes a period where the resultant effects of gestures tend 
to make them more specific, for example, the hand which enters the 
visual field and obscures the view, or the sound produced by certain 
muscular contractions of the vocal organs. This association must 
mark the moment when the fields of kinaesthetic, visual and auditory 
etc., imagery fuse together. Then what Thorndike calls the law of 
effect, or what Baldwin had described as circular reaction, comes into 
play. It is an important factor in the psychomotor development of 
the child. The effect obtained excites the movements which have 
produced it, and inversely. All kinds of learning processes follow 
(vocal noises and language phonemes, autopalpation and recognition 
of body activity, etc.). This period belongs to the second half of the 
first year. 

(4) Movements Towards External Objectives 

The first meetings with external objects seem to cause only reflexes 
without relation to the objects themselves. For example, the hand 
which closes over whatever touches the palm does not show prehen- 
sion but rather a clinging contraction which is closer to the actions of 
holding on or climbing than those of picking, seizing, or grasping. 
The period when the child becomes capable of reacting to an object 
as such appears towards eight or ten months. It is the period of 'near 
space' (W. Stern), a period of exteroceptive and exterofective explora- 
tion, of catabohc elaborations rather than anaboUc, such as those 
which could be called autoplastic and which are directed towards the 
construction of the individual, not only of his organic basis but also 
of everything which can give him shape and form. 

Motor activity can develop in two different orientations which are, 
moreover, interdependent: one, whose source lies in the sphere of 


attitudes, that is to say in tonic activity, but wliose later develop- 
ments can use all the other forms of motor activity or mental 
activity (imitation, for example); the other could be said to be extra- 
verted, seeking its effects in the outer world, but requiring the first in 
order to become precise and consolidated. A large number of 
activities can be considered alternately from one or the other point 
of view; for example, motor habits, according to whether they are 
related to their object or to their learning. 

(5) Motor Stages and Syndromes 

The successive stages of motor development, which moreover 
overlap each other, can then be distinguished as the affective-motor 
stage, the sensori-motor stage, and the objective-motor stage, which 
could also be called the projective stage, since the perception of an 
object calls for a total projection of activity, the perceptive or con- 
ceptual settings remaining indistinct or inert. In every motor act, 
however, intervene neurological components, which can be more or 
less retarded or deficient, one or the other. As a result certain func- 
tional deficiencies or fragiUties occur, which lead to different syn- 
dromes or motor types, each in connexion with the corresponding 
psychic dispositions. 

The regulation of tonus can be affected in different ways. Some 
children in the larval state, and depending on the occasion, show the 
stiffness which is observed in Parkinson's disease. In others, of a 
quite different psychic type, the hypertony is shown by contractions 
of the face or the body such as occur in certain pallidal lesions. They 
are mobile and can eventually give way to generalized muscular 

There also exist cases which closely resemble cerebellar asynergy. 
Another syndrome comprises muscular instabiHty accompanied by 
tremblings and very slight irregular displacements of the head, 
shoulders, trunk, and limbs, which if enlarged would give a picture 
of chorea and which I have called for that reason subchoreic instabihty. 

Finally, there are various syndromes which appear to refer directly 
to the functions of the cortex. It seems possible to localize two of 
them fairly accurately. One of them is basic to apraxia. It concerns 
the global comprehension of the motor act and its carrying out 
according to well arranged phases. The other affects that field of 
psychic activity which is closely related to attitude and particularly 
with those reactions which can be aroused in us by the presence of 
another person and which I have called reactions of bearing. The 
syndrome would be one of prefrontal inadequacy. 

This brief outline has, of course, not taken into account the many 
details whose description here would lead to useful conclusions or 


confirmations. It is only intended to give the main lines of the func- 
tional plan with which psychomotor activity complies. 

I will not dwell on Professor Wallon's system. I shall simply try to 
show where the Piaget (1951)-Wallon (1947) controversy lies as 
regards these ideas. 

Personally I do not beUeve that the opposition between Piaget and 
Wallon, which some have wished to estabhsh and maintain, exists. I 
have the impression that they have each put themselves in a different 
perspective and have considered things according to their own 
temperaments. Moreover they illustrate very well how, starting from 
the same estabhshed facts, divergent interpretations can be evolved, 
which are not, however, necessarily in opposition. 

The two aspects which seem to me different are the following : 

Firstly, with Piaget (1936, 1947), as far as the infantile stage is 
concerned, the interest is focused mainly on the genesis of sensori- 
motor intelligence. For him the essential is to get a grasp of the very 
first elements in this genesis, even before the appearance of sensori- 
motor intelligence. He searches for the most fugitive signs in the very 
first weeks of life. 

Wallon, on the other hand, is preoccupied with something entirely 
different. He studies by his own dialectical method how, starting 
from the same organic sources, very different types of behaviour 
separate out. Thus, deriving from the motor activity he observes 
between the ages of three and six months, he finds motricity orienting 
in two different directions. The first is an aspect of motricity which 
is in a sense made of tonic material and constitutes a starting-point 
for affectivity and for syncretic affective sociability. The second 
orientation is the sensori-motor sequences, those circular reactions 
which will be the starting-point for the idea of object and for repre- 
sentation. Wallon attempts to show how these two aspects of human 
behaviour, starting from the same organic sources, finally determine 
each other. 

The second difference, or at least apparent difference, between the 
two systems concerns just this idea of syncretic sociability. This is 
where Piaget (1926) and Wallon (1951) have been most strongly 
placed in opposition to each other by those who in my opinion have 
schematized both. 

Generally Piaget is schematized thus : the child starts from a sort 
of autism — moreover a relationship is established between Piaget's 
autism and psychoanalytic autism — and gradually the child becomes 
a social being. So, starting from an egocentric state — and even, going 
farther back in childhood, one might speak of a sort of solipsism — 
from which the child at about six years socializes himself until at 
about twelve years he achieves the feeling and idea of reciprocity. 


According to Wallon there can be no question of autism in early 
childhood— unless, of course, the term is redefined. Wallon insists 
on the extreme sociability of the small child and says that probably 
at about six months the child is the most sociable of beings, having 
that syncretic sociability which he has defined. Then evolution will 
not be from a state of autism, of withdrawal into oneself — which by 
the way Piaget never said — to a social state : progress will start from 
a state of non-differentiation, a state of communion with the mother, 
and will gradually reach a definition of interdependence between self 
and others. At the same time as this evolution is taking place on the 
affective plane, sociability is taking on another aspect : it is becoming 
more intellectual, while continuing to retain this affective, emotive 
basis, which can be found still even in the adult. 

There is, therefore, no opposition between these two authors, but 
they treat different functional levels of sociability. Personally I regret 
that the words 'autism' and 'egocentrism', used by Piaget with all 
sorts of precautions, should have given rise among careless readers to 
false interpretations. 

Main Stages of Development from Three to Eighteen Years 

I should now like to pass on to the second part of my communica- 
tion, that is to the evolution of the child between three and eighteen 
years. Rather than giving a complete and dogmatic exposition of this 
problem I will refer to certain stages which we might discuss. 

I shall now abandon the frame of reference used by Wallon the 
neuropsychiatrist and neuropsychologist and use one which contains 
nothing scientific: scholastic organization as it exists in most of our 
countries. Whatever your opinion on the value of scholastic pro- 
grammes and pedagogic methods, you will no doubt admit that the 
school must take into account, at least approximately, the child's 
possibilities, and that the main scholastic stages schematically inter- 
pret a sort of implicit psychology. Moreover, historically the question 
of mental age originated from the question of school age. Age 
differentiation became useful only when compulsory education was 
instituted in most of our countries. The identification of the stages 
and forms of intelligence became more precise with the diversification 
of studies as formulated, for example, in the British Education Act 
of 1944 and the French 'Plan de Reforme' of 1947, known by the 
names of its chief promoters, Langevin and Wallon. 

To come back to the scholastic stages, let us see how the psycho- 
logists have defined their significance and their limits. 

Between three and six years there is a pre-school stage in almost 
every country. Attendance is optional, and there is not so much 
instruction as education — at least that is so in theory. 


From six to twelve years comes the stage of primary schooling, 
which consists in the acquisition of the basic intellectual automatisms 
reading, writing, and counting. 

From twelve to about eighteen comes the stage of so-called secon- 
dary schooling which culminates in the matriculation (baccalaureat) 
type of examination. 

Beyond eighteen years comes the post-school educational stage or 
university studies. 

It is interesting to note that these ages, three, six, twelve, eighteen 
years are the same in most countries with a variation of about one 
year. It is also interesting to consider that reformers, when demanding 
an extension of schooUng beyond twelve years, no doubt take into 
account reasons of a social nature. They want to democratize the 
school, open the school to the whole of the population and give 
equal opportunities to all children. But according to the most recent 
demands for reform, we see that another argument is invoked: in 
actual fact the intellectual evolution of the child is not finished at 
twelve years, and to achieve a harmonious development of the 
individual instruction should be prolonged until fourteen or sixteen 
years, perhaps even to eighteen years. It is instructive in that respect 
to read the 'whereas' clauses of the British and French plans for 

School practice has not only prolonged the hmit of studies but has 
also determined the sub-divisions within the main primary and 
secondary stages. By and large a sub-division is marked at nine years, 
when the mechanisms of fluent reading and writing have already been 
acquired (end of elementary course according to French terminology). 
Another sub-division is made at about fifteen to sixteen years, as 
shown, for example, in the two stages of American secondary educa- 
tion (junior high school and senior high school) and the two stages 
of the French grammar school (orientation stage and determination 

It seems then that the school in our different countries has marked 
tJie following stages: three, six, nine, twelve, fifteen, and eighteen 
years. It is to be noted that there is some kind of concordance of 
psychological studies with these different stages. 

As for three years, for a long time psychologists have marked this 
as a period of crisis and of socialization accompanied, moreover, by 
an attitude of opposition. The child experiences a certain confusion 
and uneasiness and yet new possibiHties of the beginning of a collec- 
tive Hfe are appearing. The nursery school or kindergarten answers 
the needs of this age. 

At six years an emotional stability is seen to appear in the child. 
The child enters the phase called by psychoanalysts the latency 


period which, according to other authors, is a period of objective 
interests. One fact we can definitely estabhsh is that the child becomes 
capable of fixing his attention and concentrating for fifteen or twenty 
minutes on exercises he would have found very boring previously. 
Dr. Grey Walter mentioned the same phenomenon in his communica- 
tion: at about this age the E.E.G. takes on a fairly characteristic 
form. The alpha waves are dominant. We have been able to ascertain 
by using a very old test, the crossing-out test, that the child of five 
years is quite incapable of this eflTort. For about ten years we have 
been using a rather special technique which consists in giving two 
degrees of difficulty in this crossing-out test : in the first, one sign has 
to be crossed out; in the second, two signs have to be crossed out. 
Thus the reactions of a child faced with two degrees of difficulty can 
be compared. At five years the child is incapable of carrying out this 
test. He abandons it very quickly. At six years the task becomes 
possible; 60 to 70 per cent of the children are capable of performing 
it. At seven years the adult formula is already established. At six 
years the formula for behaviour is very characteristic: the child's 
activity, although well adapted when he has to discriminate a single 
sign, tends to become disorganized when there are two signs. 

From the age of seven years the adult formulation is reached, 
though naturally with a diff"erence in speed. I think that this shows an 
essential phenomenon: the possibility of concentrating on tasks 
which do not involve the immediate interest of action. The child is 
capable of sustaining his interest in order to act on an intellectual 

At about twelve years a very remarkable fact is noted: the most 
diverse psychological methods are unanimous in describing this age 
as a sort of culmination. Beyond twelve to thirteen years the tests are 
hardly discriminatory, or rather they are no longer discriminatory as 
tests of development. It would seem that individual differentiations 
are then more apparent than diff"erentiations due to age. We all 
know of the work carried out during the First World War where it 
appeared that the mental age of the recruit, the ordinary soldier, 
was twelve years. This does not mean that our armies were composed 
of mental defectives; it means that the age of twelve marks the 
culmination of a certain type of development. The same conclusions 
are reached in Piaget's work, which shows that the principles of 
conservation have been acquired on all levels by this age. 

However, we certainly have the impression that evolution continues 
beyond twelve years. How does it continue? It is a fascinating prob- 
lem from the physiological, neurophysiological, psychological, and 
social points of view. Unfortunately our documentation on this sub- 
ject is much less sound than that on the child of zero to twelve years. 


Much has been written on adolescence but in my opinion very little 
of it is vahd in the field of psychology. One notes that the adolescent 
as described in psychological works may be the adolescent of a 
certain epoch and particularly of a certain social milieu, that is to 
say the milieu of school children and grammar-school pupils. Does 
this very romantic picture of the adolescent, with this profound 
crisis, this personal upheaval, correspond to reality ? I do not believe 
so. Of course this type of adolescent exists; I have certainly met it in 
the grammar schools. 

We have carried out certain comparative studies on adolescents 
in different environments, some already working in factories, and 
others continuing their studies. It is most surprising to discover that 
among those who have already entered a social life a hostile attitude 
towards the family is very seldom met, whereas this attitude is at its 
maximum among grammar-school pupils of the same age with the 
same material conditions, that is to say, having as much pocket 
money as the young worker. It is not, therefore, the fact of having 
money which explains the attitude of these young people, it is the 
fact of earning it or receiving it. Moreover, a much earlier autonomy 
is found in the young worker than in the young grammar-school 
pupil. Obviously this is not sufficient to give a valid differential 
picture because, even if we find here that the worker has superiority 
over the grammar-school pupil, on the other hand, the grammar- 
school pupil has the superiority in that he becomes aware of cultural 

Our knowledge of the adolescent is still very imperfect. Perhaps 
what we understand best is what is revealed on the intellectual plane 
by the possibility, for a boy or girl of fifteen to sixteen, of attaining on 
the practical or theoretical level a hypothetico-deductive attitude. 
What does this signify ? I will make a very shaky hypothesis : at about 
twelve to thirteen years all the fundamental mechanisms have been 
acquired. Beyond that the cultural element comes into play much 
more than the developmental elements, although evolution continues 
on a physiological level, but the mechanisms already acquired at 
twelve to thirteen years still need to be made flexible, through exer- 
cise. In this connexion Simon (1939) said that it was not sufficient to 
have intelligence, it was necessary to know how to use it. Perhaps it is 
just after twelve years that this intelhgence begins to be used in 
different directions. I mean by that that psychologists are perhaps 
wrong in looking only for the limit of intelhgence. The age of twelve 
to thirteen years marks perhaps a new departure. Intelligence is 
nothing if it is not creative and intelligence can only be effective if 
it goes back to certain aff'ective sources. Now it is perhaps during 
this period from fourteen to eighteen years that the human being, 


coming into contact witli new social, human, and affective realities, 
manages to give a deep, concrete sense to all the perceptive-intellectual 
mechanisms which he has acquired during the scholastic period. 

In short, I wished to underline our lack of certainty about this 
period of adolescence, on which much has been written. This study 
has been perverted because not enough account has been taken of the 
cultural and social dimensions which give direction to adolescence 
and without which this period cannot be defined. I think that beyond 
twelve years cultural factors are of such importance that it is necessary 
to proceed by a differential means in order to reach significant results. 

Evolution of Graphic Ability and Language in the Child 

I will now pass on to a statement on two series of behaviour: 
evolution of graphic ability and evolution of language. For these 
examples we shall return to early childhood. I apologize again for 
the schematic nature of this sequence but it is only intended to serve 
as a basis for discussion. 

On the subject of the evolution of graphic abihty (Ajuriaguerra 
et al., 1949) I should like to stress the fact that it seems to me to be of 
capital importance and that it starts at about the age of eighteen 
months. The child then becomes capable of copying a vertical or 
horizontal line, but he can only copy the line right next to the model 
line. We frequently observe this phenomenon of 'sticking close' in 
the child's first graphic images. This tendency seems to last up to 
two or two and a half years. 

At three years, the child becomes capable of copying a few capital 
letters but he copies as he draws; the cursive movement of writing 
appears about six months later. The child very early has a different 
attitude to drawing from writing. 

At four years he is capable of copying words but he puts them any- 
where on the paper. At the beginning he writes at the bottom of the 
page and at about six years, under the influence of instruction, he 
begins to write at the top on the left. 

At five years he can copy phrases and becomes capable of writing a 
few words spontaneously, but with many inversions in letter order. 

At six years we have observed a very interesting phenomenon : the 
appearance of the human profile turned towards the left. This phe- 
nomenon has given rise to quite a number of interpretations. In any 
case on the intellectual plane this shows fairly considerable progress. 

Up to five to six years graphic ability is determined by motor 
abihty. The dominant hand develops less rapidly than the other with 
a retardation of one stage. I should like to recall here the law pro- 
pounded by Gesell (1945), the law of reciprocal overlapping, 
according to which the definitive formula on the motor plane is not 


established gradually in the child but by means of a play of alterna- 
tion. These alternations, according to Gesell, are the sign of a deeper 
alternation due to myelinization » 

Beyond six years the intellectual and affective factors are decisively 
added to the motor factors in the evolution of writing and drawing. 
A number of authors have studied the psychology of drawing, notably 
LuQUET (1913), who distinguished different stages as, for example, 
the passage from what he calls intellectual reahsm — a stage where 
the child draws with an effect of transparency and of skew — to visual 
realism, which is attained at about nine or ten years. 

At nine years orientation of graphism towards the left is noted 
(sinistrog3n-ation) not only in writing, but also in drawing. This 
originates as a motor effect and attains a sensory form. 

In this connexion I carried out the following experiment (Zazzo, 
1950). A child is given some very diverse figures and is asked to dis- 
cover profiles in these pictures, which are shown very rapidly. At 
seven to eight years there is no difference, he finds as many on one 
side as on the other. At nine years almost all children see only the 
profiles which are turned to the left. We wanted to confirm this law 
in adults and we found the same phenomenon. 

In short this perceptive organization, this sensory evaluation, 
appears at nine years, and from nine years on the same percentage is 
found as among adults. I make one reservation, but at the moment I 
cannot give any definite interpretation of it: during adolescence, 
between twelve and fifteen years, an extraordinary reversal of these 
proportions is observed. I do not, however, think that there is a 
perceptive disorganization at this age. During this period the right 
profile takes precedence. Is this a sign of a tendency to opposition, a 
disruption on the perceptive plane due to emotive factors ? I cannot 
say. I pass this question over to the competence of those who are 
more directly concerned with the problem of aflfectivity. 

I will carry on then with the history of graphism. Among the 
various observations that have been made I find stages and ages 
which appear to me particularly significant. With the appearance of 
organized graphism, the first stage which I consider to be important 
is the disappearance of this graphic confusion which expresses on the 
graphic-motor plane one of the fundamental characteristics of 
childhood — non-differentiation. This stage comes at about two and a 
half years. 

Still another important stage is that at six and a half years when 
the child is capable of well-organized drawing and attempts to write. 

Another important age is that of nine to ten years. At this age, 
when the profile experience occurs and where it is no longer the hand 
which is responsible but the eye that guides the hand, the child is 


found capable of dissociating himself from motor conditions and 
dominating them. There is an organization of a perceptive field, a 
sort of perceptive Gestalt, whose genesis we have already seen; there 
were first neuromotor determinations, then towards nine years 
demands of a sensory nature which dominated motivity. It is at this 
age that a child's drawing attains conformity to appearance. It is also 
at this age that he attempts to read and write fluently. I think that the 
convergence of these facts is highly significant. 

Beyond ten years the story continues but the cultural factors 
markedly dominate developmental ones. It is no doubt necessary 
that intelligence should continue to evolve for the child to be capable 
of giving chromatic relief to perspective, but it is obvious that 
education is a most important factor and that a person who has 
received no instruction cannot produce perspective even if psycho- 
logically and physiologically he is capable of doing so. 

It is interesting to note that all the comparisons that have been 
made on children of different cultures show the same evolution up 
to ten years. One of our colleagues, Prudhommeau (1951) apphed 
drawing tests to different ethnic groups, Eskimos and others. Exactly 
the same stages of evolution were found. No diff"erence has been 
found between French children and children of other nationalities 
that we have been able to examine. 

I will now pass on to the problem of language. It would be very 
interesting to define the common points by comparison of these 
different lines of behaviour. 

In the first two months, there is wailing and emotive crying. 

From three months onwards we get babbling and warbling, and 
certain utterances gradually take on an expressive character. 
According to Pichon (1947) this is pre-verbahsm which has, more- 
over, been confirmed by most of the English and German authors 
and has been fixed at the age of about six months. 

Then comes the stage which certain authors have called the stage 
of pure comprehension : a child understands but cannot yet express 

The first word appears towards the end of the first year, but it is 
very difficult to fix a date for the first significant word. There is much 
discordance among authors on this point and by saying that the 
ages are considered to vary between nine and fourteen months I will 
not have added anything very interesting to the discussion. 

Are there any natural words which are found in all cultural miheux ? 
The word 'Mama' has been suggested and the negation 'no', which 
marks a movement of refusal of food. One thing is certain, that is 
that at nine months (possibly a little before) a child is capable of 
imitating certain syllables. The child has the lip movements which are 


structured into sounds or phonemes. Is the 'Mama' a continuation of 
sucking? This is possible. Has this word a definite significance? I am 
quite certain that it has not. Of course an association is gradually 
built up between the mother and this 'Mama'. Under normal con- 
ditions it is always reinforced because the father and mother watch 
out for the appearance of the first word, but when it is not reinforced 
the word does not take on this significance. 

I carried out an experiment which might seem inhuman to you. 
With my first son, who is now eighteen, I did not confirm the 
'Papa' and 'Mama' so that the child has never called us 'Papa' 
and 'Mama'. Even to the present day he calls us by our Christian 
names and this horrifies many of our friends. We admit that this is 
inconvenient. Moreover, we have seen that this experiment may be 
dangerous because when the child became aware of it for the first 
time, at the age of three, he had a shock because the teachers were 
horrified and tried to force him to call us 'mother' and 'father'. 
This he always resisted. The later children say 'Daddy' and 'Mummy'. 

After twenty months the first phrase appears. Although it is diffi- 
cult to estabhsh exactly when the first word comes, which is not very 
important in the evolution of the child, the first phrase, on the other 
hand, seems to me highly important and in direct relation with the 
general intellectual development of the child. Mentally deficient 
children will say the first word at about the same time as normal 
children, but, on the other hand, their first phrase is always delayed. 
We still have to define what is the first phrase. Most authors agree in 
considering that there is a phrase when the child is capable of uniting 
two words in a language. It is then that the true regulating function of 
language appears: the child filters his phonemes and emotive cries 
through a web of tenses and begins to organize them. 

From two years onwards the number of grammatical possibilities 
increases with an extraordinary rapidity. Functional normalizations 
start with the third year. Here again we find this a significant age. 
Prepositions, declensions, inflexions, tenses, genders, and number 
appear and there is a change from pre-grammatical language — what 
might be called Pidgin English {petit negre) — to organized language. 

By three years pronouns, past tense, and plurals have been acquired. 

By six years, language is almost complete on the verbal plane. 
We should notice here a certain time-lag : at five years the child can 
use all tenses and can even use the subjunctive properly if it has been 
used in his hearing. 

There are wide individual variations in the attainment of these 
functional standards. Naturally, if certain tenses are not used in the 
hearing of the child, he will not use them either ; but he is capable of 
imitating them and a person who imitates shows by doing so that he 


is capable of understanding a situation. At five years, then, acquisi- 
tion is practically complete. Later we shall find that there is a re- 
apprenticeship when it comes to written language; this is a new 
functional level and is of considerable interest because with written 
language symbolism in the second degree appears. Spoken language 
is already a symbolism because the word symbolizes the thing. 
Written language symbolizes verbal language which already itself 
symbolizes the thing. 

It would have been interesting, in order to understand the mecha- 
nism of language evolution better, to consider the sequences within 
the different grammatical categories; but we have not enough time. 
I will simply recall that 'no' appears before 'yes' and I think this is so 
for all languages. This is no doubt due to the fact that 'no' has a 
much greater aff'ective value than 'yes'. Adjectives appear before 
adverbs, and adverbs of place before adverbs of time. 'Why' — at least 
in French, though possibly not in all languages — appears before 
'how'. 'Me', which is more affective, appears before T. The question 
of the acquisition of pronouns is worth a study in itself, since it is 
with the mechanism of pronouns that the child expresses which 
stage of decentralization he has reached, because the pronoun changes 
according to the person who is speaking. This is a fundamental 
acquisition. Here again it is necessary to distinguish between lan- 
guages because this acquisition varies with the language structure. 

It would be worth comparing this language evolution with the 
diff"erent stages of grapho-motricity and with other sectors of 
behaviour. I think, in fact, that it is up to the psychologist not to 
isolate one sector but to see how the progressive differentiations 
occur and to discover, despite possible overlappings, certain 
functional levels. 

We have tried to compare the different aspects of child behaviour 
which could show the attainment of awareness of self (Zazzo, 1948), 
I will only indicate one aspect of our studies, which was the first 
chronologically. We studied the evolution of language in a young 
child, particularly of personal pronouns as related to the recogni- 
tion of his own image in three different situations: in the mirror, 
on photos, and in films. From this study we obtained some very 
interesting data. Recognition in the mirror, on photos, and in 
films is very much delayed when the precaution is taken, in the case 
of the mirror, of not giving the child any suggestive help or any 
intensive training. We find that the recognition does not occur until 
two years and two or three months. It is also found that the recogni- 
tion of self and of others is interdependent — the recognition of 
others coming, however, slightly earlier — and this is so for the three 
situations. With the mirror it is very clear. I have noticed that 


although the child is no doubt looking at himself in the mirror he 
concentrates on the image of the other person because he knows the 
other person whereas he has never seen himself. There are games in 
front of the mirror where the child looks behind the mirror as an 
animal would. Then at about two years we have noticed there is a 
kind of disorganization as if a sudden state of awareness of self had 
caused an affective upset. A few weeks afterwards the child recognizes 
himself in the mirror (at about two years, two months). Up to the 
end of the third year he displays a certain anxiety and at the same 
time a certain pleasure in looking at himself. At about two years, 
ten months this disappears; the image has become famihar and no 
longer causes uneasiness. However, at two years he recognizes other 
people on photographs but does not recognize himself until at least 
six months later. 

As for films, he recognizes the situations in which the film was 
taken; he recognizes the beach or the room where he was, but he 
does not recognize himself. At two years he recognizes other people 
without hesitation but he himself is a child whom he does not know. 
When he does recognize himself at about two and a half years, he 
uses rather a strange form of speech, as if there were a duplication 
between himself and the image. He says 'there are Johns' or 'there is a 
John'. He speaks of himself in the third person. 

These phenomena have been confirmed with other children. They 
always occur during the third year though the exact age varies, and 
the recognition of other people always comes before the recognition 
of self. 

Now this follows very closely the evolution of language. It is at 
about three years that the use of T appears without hesitation and 
is used grammatically. Before that there are no doubt syncretic Ts' — 
he says 'I don't know' — but the T is only properly distinguished at 
about two years ten months. On the other hand, the grammatical and 
possessive forms 'me', 'you', and 'your' appear simultaneously. The 
personality crisis with opposition and negativism appears in more or 
less crude forms between two and a half and three years. 


Is there any difference between the recognition age for black and 
white photographs, which are an artificial convention, and colour 
photographs ? 


I have not carried out any systematic experiments. I have tried 
with colour films and the results appear to be about the same, but I 
cannot give a definite reply here. 



Now I should like to draw some conclusions on the laws of child 
evolution and make some observations on method. 

To come back to Gesell — whose work I have drawn upon several 
times during this communication, without naming him — growth is a 
unifying concept. From the beginning Gesell (1929, 1945) rejects 
dualism which, according to him, would make us incapable of 
properly understanding the liaison between the physiological sub- 
stratum and behaviour. This can be seen, moreover, in the structure 
of his tests, where there is no duality, but in any case a differentiation 
occurs in proportion to age. This is a law of psychological evolution 
and also of neuro-physiological evolution. 

The first general law then would be a law of differentiation and of 
progressive integration apparent on all levels. 

Firstly there is differentiation between various intellectual activities. 
The studies of Pieron (1949) and of his pupils (Fessard, 1931; 
MoNNiN, 1933), which dealt with scholastic and professional orienta- 
tion, came to the following conclusions: the correlations between 
different forms of intelligence are very high at about six to seven 
years and become gradually lower as the child grows. There are both 
differentiation and integration at the same time. 

On the affective plane as well, we find during childhood a differen- 
tiation occurring between self and others, a differentiation which is, 
at the same time, an interdependence, and which the Piaget school 
calls reciprocity. 

Thus on all planes, differentiation and integration lead the child 
to finer and more flexible adaptations and to an autonomy which is 
not an isolation but is, on the contrary, an interdependence and an 
organization. One could formulate this law in a very sketchy way by 
using the terms of child psychologists as follows : the child evolves 
from syncretism and from non-differentiation to a synthetic activity 
or on the motor plane to synergies and re-organizations of movement. 


Dr. Zazzo mentioned the behaviour of children when confronted 
with a mirror and its use as a test for self-recognition. Now in some 
of our machines we have demonstrated, quite accidentally, precisely 
the same type of behaviour. There are certain simple machines which 
display a characteristic mode of action when they are presented with 
their own reflections in a looking-glass. It is quite unique, an always- 
and-only response, so that it is diagnostic, in a zoological sense, of 
self-recognition. These same machines, when confronted by one 
another, again display a unique but different behaviour diagnostic of 


social recognition; they build up a society which has characteristic 
modes which can then be interrupted by a common stimulus. The 
social organization can be broken down into a competitive complex, 
possibly with the destruction of all the machines. These two behaviour 
modes of recognition of self and of society can be analysed into quite 
straightforward cybernetic ciphers. One can describe these processes 
as being reflexive — I use that word deliberately — and the appearance 
of them depends upon certain well-defined or definable mechanisms 
appearing within each individual machine. One can say that if these 
mechanisms appear, reflexive behaviour will start, it will have its 
diagnostic characters, and if observing from outside one would say: 
'this is a recognition of self, or : 'at this stage a recognition of society 
appears'. I think this sort of application of cybernetics to these 
problems can be enormously fertile. It doesn't deny anything (and 
that's a very important point) but it can affirm a great deal. 

I would like to ask Dr. Zazzo one rather trivial question. The use 
of the term reflexive was quite deliberate on my part — as far as I 
know it is my own particular use of this word; I don't use retroactive, 
or feedback, but reflexive. Now does Dr. Zazzo know whether the 
attitude to self in the child — the time at which it develops, the way 
it develops, the form it takes — is related at all to the linguistics of the 
child? In English you teU the child to say 'I wash' or 'I am washing', 
in French you have to say 'Je me lave' — I wash myself. That is a 
reflexive verb and in French the reflexive is used a great deal more 
than the passive. 


In Russian you haven't got a passive, you just use the reflexive. 


It often seems quite ridiculous to English people to talk in French 
about an inanimate object doing things to itself, finding itself, etc. 
Now, have you any evidence that that affects the way in which a child 
develops its attitude to itself? 


At the age when the child recognizes himself in front of the mirror 
his language is not yet sufficiently well organized for one to be able 
to see a transformation occurring in the reflexive verbs. From the 
linguistic point of view we cannot have any indication because these 
verbs do not exist for the child at that age, that is at about two years 
two months. On the other hand, we have been struck by the fact 
that the child begins at this period to use personal and possessive 

12 177 

pronouns and 'me' and 'you' simultaneously. This reply is not very 
satisfactory because the reflexion you speak of cannot appear. 

Your comparison with machines is very interesting, and raises all 
kinds of problems. How far is the analogy legitimate ? In the case of 
machines, would there not rather be an analogy with mimicry, a 
reciprocal imitation ? What appears to me to be characteristic in the 
child's behaviour in front of the mirror is his uneasiness. There is a 
sudden awareness, a hesitation; he sees another and starts to be 
aware of himself. It is only after several months that this uneasiness 
disappears and he looks at his image naturally. I see in this uneasiness 
the mark of a sudden awareness of self which could not have any 
analogy with the example you gave. 


The disturbance and anxiety of the child when it recognizes itself 
in the mirror is perhaps based on the fact that somehow the child 
feels 'Well, I am here, I cannot be there too'. It is really a very diffi- 
cult thing to understand that my being here doesn't exclude my being 
there, though it doesn't seem so for adults. But birds do not know, 
and neither do dogs, that one subject cannot be in two places at once. 
That's very surprising to us, because to us it is self-evident that an 
individual object which is here cannot be there. But I know of two 
chance observations, one in dogs, and one in my pet greylag goose 
Martina, which show quite clearly that this self-evidence doesn't 
exist for the animal. I was coming back from a long canoe trip with 
Martina, and I was dressing a.nd Martina was preening, after a long 
paddling tour on the Danube. Suddenly Martina gave her flight call 
and prepared to take off", which was very disagreeable to me, because 
I wanted to go home and didn't want to lose her. She called again and 
again, her social caU, which was at that time directed exclusively to 
me, and then I saw what she was seeing. She saw myself, a man in an 
identical canoe on the other side of the Danube — a fat fellow with a 
beard. I instantly realized what she was thinking, and tried attracting 
her attention, saying 'Hullo, here I am' but I couldn't prevent her 
from taking off. She flew over the Danube and landed in front of that 
man, and then she recognized that it wasn't me, and go so frightened 
that she rocketed like a firework, and came back to me for protection, 
then greeted me violently and threatened in the direction of that man. 

The second instance is a dog which Heinroth had when a student, a 
kind of GriflFon bitch. Heinroth had a girl friend, and when he went 
for a rendezvous, the dog knew it, and smelt every lady that looked 
approximately hke her in order to greet her when she saw her. Now 
when she had come, and the dog had greeted her, this didn't prevent 


the dog from looking for other girls resembling her. She wanted 
another individual of the same kind. 


I have a rather detailed study of a child working toward this 
recognition with films, because I have persistently shown my child 
films of herself, and I started quite early to watch what happened. 
She went through the stage that Dr. Zazzo described, in which she 
would call the child in the film first 'the baby' and then 'that Cather- 
ine', which was different from 'this Catherine', and then came 
the moment when she sorted the thing out. Watching a picture of 
herself swimming, she was standing with her hand on her doll's 
perambulator and she looked at the picture and said 'That's 
Catherine, I'm swimming, that's Catherine swimming', and then she 
shook the doll carriage, then took up the swimming position, and 
then said, '/'m shaking the doll carriage, that's Catherine swimming. 
I'm shaking the doll carriage', and made the time distinction. This 
was at about two years and nine months. 


I should like to spotlight this phenomenon by means of a reverse 
phenomenon of a pathological kind. I think what is essential here is 
the formation of the image of the body itself with visual, postural, 
etc., elements: an image which, in my opinion, is closely associated 
with the genesis of awareness of self. In certain kinds of pathology 
disintegration is often the reverse phenomenon of doubling, of 
positive or negative autoscopy. I am thinking of the case of Guy de 
Maupassant. The sick man loses his image, or, on the other hand, he 
may have an hallucination. He sees his own image appearing in front 
of him. This is a phenomenon which has been frequently underhned 
in psychiatric literature — and also in other literature. It is a theme 
which is used, for example, in The Student of Prague. We have then 
two sides to this phenomenon : the establishment of the image of self 
which is absolutely necessary for the awareness of self, and in certain 
pathological cases a disintegration where the individual sees himself; 
one might even allude here to Socrates' demon. 


I have followed with great pleasure M. Zazzo's communication, 
which testifies to a most discriminating and concihatory attitude. 
M. Zazzo said that the Wallon school is often placed in opposition 
to the Piaget school. I think, as he does, that this opposition is partly 
the result of a false interpretation of their ideas given by the pupils of 


these two teachers. However, M. Zazzo knows as well as I that these 
two creators of systems can discuss their ideas in all cordiality and 
friendship, of course, but it is evident that every creator of ideas 
experiences some difficulty in putting himself precisely in the place of 
another because above all he defends his own creation. We, the 
followers, have not the same reasons for being aggressive or egocen- 
tric. We are quite naturally drawn to look for means of unification 
rather than of separation. 

I should now like to take up again certain distinctions estabhshed 
by M. Zazzo between the two trends of thought and to add to them 

Firstly, there is a difference in the points of view and the problems 
raised: M. Wallon raises the question of the origins of all psycho- 
logical behaviour. This orientation of mind is shown by the titles of 
some of his works : Les Origines du caractere chez V enfant {The origins 
of character in the child) (1933); Les Origines de la pens ee chez V enfant 
(The origins of thought in the child) (1945). M. Piaget is more interested 
in the end of evolution. He studies La Genese du nombre {The genesis 
of number) (Piaget and Szeminska, 1941); Le Developpement de la 
notion du temps {The development of the concept of time) (Piaget, 
1946a); etc. (Piaget, 1946b). 

Next there is a difference in method : M. Wallon seeks, above all, to 
study the complexity and the clash — frequently tumultuous — of the 
many factors arising during a phase of development or the passage 
from one phase to the next. M. Piaget attempts rather to determine 
the most general comprehensive laws which he interprets as laws of 
equilibrium. The fact that these methods are complementary is 
shown in the choice of the facts studied. 

The results themselves are complementary. I consider that M. 
Wahon has made an important discovery which M. Zazzo has per- 
haps not emphasized sufficiently, that is, the positive role of emotion 
(Wallon, 1942). Emotion is often supposed to play the negative role 
of a disturbing agent. Now, M. Wallon has demonstrated the very 
positive role played by emotion in mental development, facilitating 
even the formation of thought. M. Piaget, on the other hand, has 
neglected the emotive aspect of behaviour; he does not deny it, but 
he has not dwelt upon it becapse he has attempted above all to 
determine and circumscribe the structures of adaptation to reality. 

Let us now come to the controversy on egocentrism which M. 
Zazzo underlined. You perhaps know that the discovery of egocentrism 
on the one hand made the reputation of M. Piaget (1923) and on the 
other hand appeared as an obstacle to the understanding of his later 
work. Personally, I think that the term 'egocentrism' is not particu- 
larly suitable. Actually, M. Piaget has attempted to make it more 


precise. At present I think lie would uphold the following point of 
view: the social does not exist in itself but as a series of social ex- 
changes maintained by the child with his environment. M. Wallon 
expressed something analogous by using (as did Claparede and 
Piaget) the term 'syncretism' which clearly expresses the initial state 
of fusion existing between the ego and the non-ego. Perhaps M. 
Piaget did not manage to make himself understood when he called 
the first social exchanges egocentric relations, establishing an analogy 
with autism in Bleuler's sense, but it seems to me that the essential 
point is to show how this first form of exchange is gradually modified 
until relations of reciprocity and co-operation are achieved. Lack of 
co-operation due to incapabihty of seeing things from another's 
point of view does not in the least preclude fusion with another or 
imitation of the model. Confusion with its intervening mechanisms 
of projection and identification is, in fact, an obstacle to true co- 
operation. Thus, here again it is not actually a question of discord, 
but rather of complementary interpretation of the same facts. 

At the moment I can only see one point on which there might be 
divergence: the passage from sensori-motor activity to the activity 
of thought (and we should have to go deeper in our research to 
discover how much is fact and how much is interpretation). 

According to M. Piaget there is a continuous transition between 
the sensori-motor behaviour of the infant and thought as it develops 
during early childhood. M. Piaget tries to show through precise 
experiments (among others, stereognostic tests : Piaget and Inhelder, 
1948) that thought has motor origins. Mental representation is 
prepared by a whole number of sensori-motor explorations (the 
mental image being partly shaped by previous tactile explorations). 
On the other hand, M. Wallon (1942) insists on the opposition 
between these two phases of mental life. According to him, thought 
incorporates into itself new elements through contributions from 
language and social life in general, but this is a matter of hypotheses 
requiring verification. 

Finally, it is interesting to note that Wallon and Piaget seem to 
have travelled in opposite directions. Each of them has modified the 
orientation of his research. M. Piaget began by studying the child in 
his social manifestations, such as verbal exchange. This led up to the 
study of the child's sensori-motor activities, which M. Piaget is 
trying to relate to the nervous structures themselves. M. Wallon 
(1925), taking the opposite path, centred his early work on motor 
development, whereas now he is concentrating more particularly on 
the conditioning of the child by the social environment. 

This again is a question of complementary evolution and not a 
taking-up of unshakeable positions. 



Psychoanalytic Instinct Theory 


For my remarks I have selected out of the whole field of psycho- 
analysis the topic of instinct because I feel that instinct is the central 
core of psychoanalysis, and that the study of instinct distinguishes 
psychoanalysis from other branches of psychology, which commonly 
study other aspects of the human organism. Psychoanalysis regards 
a great deal of psychiatric illness as being due to a disorganization of 
instinctual life. I share that view, and I think that by looking at it 
that way we may prevent much mental illness. 

Freud made several attempts to formulate a theory of instinct. 
Both his main formulations were in terms of a dichotomy: 

(a) The sexual and the ego instincts {circa 1910). 

(b) The hfe and death instincts (circa 1920). 

A further important proposition was that the goal of the organism 
was to obtain pleasure and avoid pain. Freud himself did not feel very 
satisfied with his work in this field and in 1915 remarked that it 
would probably prove necessary for psychologists to look to biology 
for an adequate theory of instinct. As it happens, in the same paper 
he outlined a theory of instinct almost identical in principle with 
that of modern ethology. He conceived of instinct as having a source 
in somatic process, an impetus or force, an aim — that of 'abolishing 
the condition of stimulation in the source of the instinct' — and an 
object, namely 'that in and through which it can achieve its aim'. 

Psychoanalysts have varied a great deal in regard to which of 
Freud's theories they have adopted. In Great Britain, under the 
influence of Melanie Klein (1948), there has been an emphasis on the 
object-seeking nature of instinct. Fairbairn (1952) has supported this 
view and argued explicitly against the pleasure-pain theory. 

Freud also called attention to the fact that human beings are 
organisms which at times are driven by forces within themselves 
which they cannot easily control. We fall in love, we lose our 
tempers, we panic, we are possessed by forces which seem alien to 


ourselves; and, of course, there is the primitive theory of mental 
illness that people are possessed by a devil. It is these phenomena 
that I want to discuss and, just to give them a little more concreteness, 
I shall illustrate what I have in mind by referring to a couple of 
patients whom I have seen very recently. One is an adult of about 
forty, a woman with a severe degree of illness, tremendous phobias, 
great hatreds, deep depressions, who is subacutely murderous and 
suicidal. I have been treating her now for two and half years and her 
relations to me are characteristically ambivalent. I have experienced 
long sessions when she was shrieking at me for two or three hours 
without a break. There have been other occasions when her endear- 
ments have been on the same scale. I recall that when I came here in 
June last, on the one hand she was very sad that I should leave her, 
but, on the other hand, as she told me afterwards, she hoped I would 
drown myself in the lake — and she meant it. The thing that has 
struck me so particularly about her is that one day she can be friendly 
and cordial and we have what I call a sober session, and the next day 
she rants and raves at me as though we had never had any other 
relationship. Another feature about her is that she is terrified of 
friendly relations. She is 'happy' only when she is on angry terms 
with someone ; her sense of stability is better when there is mutual 
anger. Obviously the problem is, why is she like this? My own view 
is that this is the result of the relations which she had with her 
mother when she was small, having been extremely unhappy and 
chaotic, but, naturally, I can't prove this. What one has now is a 
first-hand and mammoth demonstration of ambivalence : in her case 
this historical reconstruction must remain very treacherous and 

The other patient is a little girl of eight. She was rather an attrac- 
tive, neat, pretty little girl. I asked her whether she knew why she 
had come to the clinic, and she said, 'Well, I have tempers, but I am 
learning to control them'. We talked a bit about her tempers, and we 
got on to her day-dreams; she imagined various things to herself. 
Sometimes she sees faces on trees — nasty faces — but she turns them 
into nice faces. Then it appears 'the nasty faces — they want to kill my 
mother'. From this I had no doubt that she wanted to kill her mother 
— or rather, a bit of her wanted to kill her mother. It's a good family, 
her mother is a decent woman, but she has tried to smack the tempers 
out of this child. As a result the child not only wants to murder her 
mother but these impulses are split off from the rest of her personality. 

Now, I give those two illustrations just to show the nature of the 
problem which we are trying to solve. 

I want to remark on three or four psychological processes which 
may be relevant, and in doing so I shall speak in a hybrid, bastard 


language which I have come to use, which derives from both psycho- 
analysis and ethology. Why do some people develop strong impulses 
to kill, and why do they actually kill others on occasion? More 
importantly, since they usually don't kill others — why don't they? 
That takes us into the field of super-ego development. It has seemed 
to me for a long time that the controHing force, conscience or 
whatever we call it in the human being, has an instinctive root. It 
seems to me that ethology, as I understand it, gives good grounds 
for thinking this is so ; I am referring to Lorenz's observation that 
the wolves' attacking impulse is inhibited by certain behaviour in 
other wolves. 

Amongst the instincts that psychoanalysts are concerned with are 
the instinct of flight, the instinct of attack, the following response,* 
which is of great importance in the study of the young mammal, and, 
of course, sexual instincts. Now, a very difficult problem is, why do 
these instinctive responses become dislocated in infancy'^ It appears 
that the instinctive hfe of the human being can become dislocated and 
that it is in infancy and early childhood that dislocation is particularly 
apt to occur. It would appear that in the early months of life the 
infant responds to sign-stimuli which are isolated in the sense that 
the infant doesn't know that they belong to an 'object' in the Piaget 
sense of the term. Spitz' work on the smiling response (Spitz and 
Wolf, 1946) suggests that the infant responds in a friendly way to 
a particular sort of sign-stimulus ; presumably his rejecting responses 
are evoked by something else. Melanie Klein (1948) has developed 
the idea of the 'good' object and the 'bad' object; I think she may be 
using the word 'object' here to mean an isolated sign-stimulus. She 
has further developed the notion that there comes a point in infant 
development where the 'good' and the 'bad' object fuse; this I 
suppose is the same thing that Professor Piaget describes as the 
'formation of the object'. It would appear from the psychoanalytic 
work that the 'formation of the object' is interfered with if there are 
very powerful and contradictory impulses evoked by the sign stimuli 
emanating from that object. Certainly, with the first patient that I 
referred to, the most striking thing about her is this split, the extra- 
ordinary independence of her hatred for me and of her affection for 
me. I can be for her either extremely 'good' or extremely 'bad' but, 
as an ordinary person, I hardly exist. 

The next problem is, if these things become dislocated in infancy 
why does the dislocation persist and affect not only the responses 

* The term used by ethologists to describe the young animal's instinct to 
attach himself to and to follow some other creature, usually the mother. Though 
it is still a controversial matter, the ethological evidence seems to me to point to 
its being an instinctive drive in its own right in the human child. 


immediately concerned — the following response, let's say, to the 
mother — but the person's sexual responses and adult love, and even 
the parental responses of a mother to her own child ? In birds at least, 
these different instinctual responses are curiously independent. Each 
one comes along, is imprinted to its object, and then another comes 
along. That doesn't tally with the human, because what psycho- 
analysts have observed is that in humans there is a horrid continuity, 
in humans the responses are not independent. Of course, one possible 
explanation may lie in the extraordinary fact (which may well prove 
a key for psychoanalysis) that a response can be imprinted long before 
it becomes operative. It may be that human sexual responses are 
imprinted in early childhood and not later. 

A second notion of the ethologists which strikes one is that each 
instinctual response waxes and then wanes. Each response, ontogeneti- 
cally speaking, has its own time at the wicket, but in certain circum- 
stances, it would appear that the batsman can refuse to leave the 
crease ! For instance, it looks as though the following response which 
waxes in the early months of life reaches its peak in the second and 
third years, and then wanes — and it looks as though that response 
can, in certain circumstances, persist into adult life. 

A third clue to the solution of this problem may lie in the fact that 
sexual responses can make use of certain components of the parent- 
child response. It is known that in certain species of bird the male's 
behaviour towards the female is similar to that of the parent feeding 
the chick, and the female's behaviour is similar to that of the chick 
being fed (Armstrong, 1942). If there are certain components in 
human sexual response which are derived from parent-child responses 
(which seems clearly to be true) it would not be surprising if problems 
that had started in the one situation should be carried over to the 

However, there are other areas that we haven't explored at this 
conference which may be relevant to this problem, in particular the 
special properties of responses learnt under stress. The two charac- 
teristics of such responses that strike me as being interesting are first 
that the response is often, in terms of adaptation, an irrelevant 
response, and, second, that once learnt it has an extraordinarily 
persistent quality. It looks as though the infant under stress can learn 
'silly' responses and the adult under stress can revert to such. 

I look upon research on the effects of separating the young child 
from his mother as being the study of the following response in 
health and disease, with particular reference to its ontogeny. We ask : 
in what circumstances is it evoked; in what circumstances does it 
■ persist; in what circumstances is it never evoked; in what circum- 
stances is it cut out, or repressed, to use the psychoanalytical term ? 


Further, what part do hereditary differences between individual 
children play in these things ? These are our research problems. 

Further, we are convinced that in research of this kind nothing but 
longitudinal studies can help us, because we have to have the details 
of the child's experience, which often cannot be obtained retrospec- 
tively even a month or two afterwards. We must observe the child at 
the time he is having the experience. 

Let me conclude by giving you a list of propositions to which most 
British psychoanalysts would now subscribe (some would add many 
other propositions) : 

(1) The dysfunction present in most people with 'emotional 
disorders' lies in their incapacity to make satisfying relations with 
other people and in particular with Ijiose who occupy the role of love 
objects to them (e.g. parent, sexual object, child, etc.). 

(2) The reason for their incapacity is that the love object arouses 
powerful and conflicting instincts, e.g. instincts to love and cherish 
and instincts to hurt or kill. 

(3) The disturbed person is unable to tolerate these contradictory 
instincts so that one or more are divorced from consciousness. 
One or more are also likely to be inhibited in some degree, though the 
drive to action persists (or, perhaps more accurately, remains hable 
to be re-awakened). Such repressed but active instincts manifest 
themselves in various symptoms and are experienced by the subject 
as tension, anxiety, and depression. 

(4) Such powerful and intolerable conflicts rarely if ever occur for 
the first time in the relations of the older child or adult towards his 
love objects. When present in an older person, they are the develop- 
mental successors of similar conflicts which were present in his 
relations to his first love objects, e.g. his mother or his father. 

(5) There are special features in the immature human being, 
especially in the first two or three years, which lead him (a) to be 
particularly prone to strong and contradictory impulses and (b) to be 
unable to sustain their pressure, with the result that processes such as 
repression, inhibition, displacement, projection, etc., occur readily. 
This is presumably due to physiological machinery for organization 
and integration being inadequate. 

(6) There is much debate as to why some individuals develop 
conflicts on a greater scale than others and why some individuals 
appear capable of sustaining these conflicts in consciousness and 
others cannot do so. All agree that the child's experiences of satis- 
faction or frustration in his relation to his parents are of some impor- 
tance in determining the strength and development of his impulses. 
Some give much weight to postulated differences in hereditary 
endowment; some give less. 


(7) There is a tremendous literature on the reasons why particular 
symptoms are developed by particular patients. Few of the theories 
advanced are entirely satisfactory; some are very unsatisfactory. 

(8) There is agreement that the efficacy of therapeutic measures 
depends far less on their ability to understand and deal with symptoms 
than on their ability to deal with the primary conflicts in object 
relations. The technique used is to permit these conflicts to develop 
within the therapeutic relationship (which they do in some fashion 
spontaneously) and to help the patient experience the impulses in 
relation to the analyst, whom he sees and feels as both a satisfying 
and a frustrating object. When these conflicting impulses are experi- 
enced together within a relationship of trust, there tends to be a 
restructuring of the instinctual Hfe, permitting less conflictful and 
therefore more satisfying relationships. 


I might give you some ideas on the value of affective factors, not 
only in neurosis but also in the development of the child and the 
social life of the adult. 

Freud and Piaget go up and down the same stream on opposite 
banks. Each remains on his own side, so they cannot meet. My main 
idea in my work is to throw bridges between the two banks and my 
object is to reconstitute the organic and functional unity of child 
development. On the one hand, there are interesting but partial 
theories, on the other is the child himself as he develops according 
to very complex laws. 

There is a need, then, firstly to study this development in all its 
aspects, and secondly to attempt to estabhsh syntheses between the 
respective data of all the studies confined to one or other aspects of 
this problem, but I will hmit myself to underlining a few points 
which seem important to me. 

As you know, Freud postulated three or four successive theories of 
instinct which are diff'erent and seem to me contradictory. It is these 
continual contradictions that have helped to disconcert not only 
biologists but also psychologists and philosophers. This fundamental 
confusion is constantly reflected in psychoanalytic jargon, and the 
reader is always constrained to examine texts most critically in order 
to know exactly what is the subject and on what plane the psycho- 
analyst is working. Thus, in psychoanalytic works the word 'Hbido' 
is used in at least five different senses, without the distinction being 
clearly indicated. It sometimes expresses sexual pleasure and some- 
times the sexual act, and aU kinds of behaviour driving the individual 
to obtain this satisfaction, etc. 


There is, then, in these texts a source of confusion between the 
energy being used, hbido or aggressivity, and the instinctual act 
which this energy is supposed to determine. Thus, according to 
Freud, instinct is an internal movement from the disagreeable 
toward the agreeable, from the painful toward the pleasant*; but 
then, according to this conception, too much insistence is laid on the 
causal factor and not enough on the intentional factor (it is precisely 
the latter, if I have understood the situation properly, which escapes 
the ethologists). Hence these interminable controversies between 
causalism and finalism of instincts. Now on this subject we can sum 
up Freud's thought as follows : an instinct is at the same time a cause, 
an object, and an aim. Here again it is necessary to distinguish 
between two very different things. According to a first theory the 
instinct is defined by a tendency to reproduce a previous state which is 
a state of relaxation or repose. The insistence then is on the actual 
energetic mechanism of this re-establishment of a previous state, 
which is a pleasant state, in order to suppress the present state, 
which is one of tension. Behaviour is then determined by the tension 
factor, and we have here an energetic conception. 

In other texts, however, behaviour is no longer determined by this 
regulating, almost cybernetic, mechanism, but by the painful, dis- 
agreeable nature of the state of tension. In fact, Freud showed that 
the final state after the relaxation is subjectively less painful than the 
initial state, hence his famous theory of the pleasure principle. 

In short, his first description corresponds to a very strict psycho- 
physiological conception; his second is a purely psychological 
approach on a subjective basis, that is to say based on sensations of 
pleasure or displeasure. The relation between these two forms of 
phenomena often remains obscure. This no doubt arises from the 
fact that psycho-analysis is at the same time objective and subjective. 

Another subject which seems to me to contain some ambiguity is 
the problem of the specificity of the object. According to certain 
theories exterior objects are never releasers of behaviour. They are 
not the actual cause of instinctive acts. They are rather the effect of 
a series of selective pieces of behaviour conditioned through heredity 
and used by the instincts to obtain their satisfaction in the exterior 
world. Thus instinct as a primary phenomenon is released by internal 
vegetative, hormonal, diencephahc, or cortical organic excitations. 
In short, drives or instincts are not initially fixed on specific objects. 
This has been demonstrated by some remarkable studies on ants 
(Brun, 1920). Entirely inadequate ('empty') anti-biological reactions 

* This psychic process is invariably started off by a painful tension and is 
directed towards the final result of a lowering of tliis tension, that is to say, 
suppressing pain and producing pleasure (Freud, 1920). 


can be released in ants. I would recall also Dr. Lorenz's experiments 
on the starling. 

It seems, then, that two separate things have been confused in 
psychoanalysis : the hereditary predisposition to a certain action and 
the instruments used by instinct either during its phylogeny or during 
its ontogeny. 

Child psychologists, when explaining the search for an object, do 
not necessarily resort to inwardly determined primary and secondary 
mechanisms. It seems that from the beginning these mechanisms are 
intimately linked and that the passage from one to the other is 
reversible. Thus one finds situations where the object itself can release 
the instinctive acts and affective reactions which define the first social 
link, this specific object being the protective mother. This is what is 
found, for example, in the 'smiling response' which takes on an 
elective character. Certain babies respond to their mother's smile 
only. The very interesting problem of the specificity of the object is 
called in question by ethologists who show that an object is not 
initially specific. In Dr. Lorenz's films we have seen geese and ducks 
following him with obvious love as if he were the father or mother. 
At the moment of imprinting, then, the specificity of the object is 
lasting and irreversible. 

Does the same thing happen in the child? Here is the great diffi- 
culty of estabhshing links between ethology and physiology on the 
one hand and child psychology on the other. I think that mother- 
fixation under normal conditions is specific. No other object can 
replace the mother without producing disturbances which are some- 
times very serious and can lead to the anxiety which is called 'the 
feeling of insecurity'. My work on what I have called 'the neurosis of 
neglect' (Odier, 1950) is centred on the origin, nature, and develop- 
ment of this feeling of insecurity in the child and the disturbances 
caused by it. It seemed to me that emotional reactions played an 
essential role in the etiology of the 'affects' and that very interesting 
results could be obtained by analysing in detail not only the instinc- 
tive causes of conflict between the libido and aggressivity but by 
analysing the affectivity itself, and by describing the phenomena not 
in terms of instinctive causality but in terms of structure. 

Affectivity, like thought, has a whole history. Affectivity presents 
very interesting structures which evolve and are entirely transformed 
when the Oedipus complex appears, when the child acquires an 
entirely new conception of social relations based on dualism. In other 
words, he relinquishes his egocentric position: he can pass from this 
well-known aff'ective participation, where there is confusion between 
himself and his mother, to a realization of a new conception, which 
for boys is identification with the father as an object. The child 


begins to feel intuitively that he himself is a subject; he can begin to 
consider things as subjective. In these entirely new relations the child 
can be the subject who experiences feelings which are either very 
ambivalent for his father or loving for his mother, and at the same 
time he may reverse the situation. It is at this moment that the 
reversibility of affective relations begins in so far as he can consider 
himself as the object and his father or mother as the subject. This is 
a very great problem. Psychoanalytic work has perhaps contributed 
to spread a certain confusion in public opinion. The Oedipus complex 
is considered as an instigator of disorder, as a necessary cause of 
neurosis. In my opinion the Oedipus complex is, on the contrary, a 
sign of health and of normal evolution. It is therefore necessary to 
explain to parents that it is a critical stage in development, but of a 
nature to ensure equilibrium in the child. 

This meeting seems to me a fine demonstration of an effort towards 
synthesis between the biological and psychological planes. I have, 
however, the impression that the dominant tendency is that of bio- 
logical reduction, that is, a tendency to carry the actual psychological 
processes to a level of mechanisms and instincts. But there is another 
method, the reverse method, which consists in passing not from the 
more complex to the simpler, but from the simpler to the more 
complex. By attempting to simplify the complex too much, one by- 
passes problems raised by the more complex. We now know, thanks 
to the work of Wallon, Piaget, and others, that the actual psychic 
processes obey not a causality — which is always very hypothetical — 
but a legality, a set of rules or laws which it would be interesting to 

What is lacking is a link between the two planes in order to avoid 
what I call the 'jumping method', where one jumps from the biological 
plane to the psychological plane without always being aware of the 
jump, and neglecting the whole territory over which one has jumped. 
It is precisely on an intermediary plane, which I call emotional or 
affective, that one can investigate what becomes of the instincts 
spoken of by Dr. Lorenz and others. What is the fate of all this 
instinctive life ? It is transformed into an affective life which leads us 
to a superior level of intellectual life. It is the affective Hfe which links 
instinctive and intellectual life. This study of the intermediary plane 
would lead us to our problem: the development, not only psycho- 
biological but also bio-psychological, of the child, and would put us 
on the level of actual experience. 


Many psychological problems have a physiological origin, and 
there is not a single psychological or philosophical thought which 


has not its correlate in the physiological sphere. Now, I should Uke 
to see the hnk between psychoanalysis, as seen introspectively and 
psychologically only, and the physiological processes. We are con- 
cerned with the psychobiological problem. I should like to see the 
same thing first with one eye, then with the other. Moreover, this is 
what we must do if we wish to advance. 


It is very hard to keep the subject to children and there are several 
things I want to ask now. First, about the characteristics of the 
imprinting mechanism, and the I.R.M. that doesn't require a reward 
or reinforcement, in relation to the psychoanalytic approach to 
instincts ; this seems to me to be of extreme importance. With regard 
to reward, I should just hke to mention an important difference in 
conditioning between appetitive reflexes and defensive reflexes. In 
the appetitive conditioned reflex, whether in the flesh or the metal, a 
reward is necessary — this is the classical association of Pavlov. In 
the defensive reflex a reward is not necessary at all. A defensive 
reflex may be self-maintained. In a machine made to imitate a 
defensive reflex I can show that a single experience may suffice, so 
that thereafter the conditioned response perpetuates itself. The 
avoidance of 'pain' or 'displeasure' is enough to make the response 
perpetuate itself as though it were imprinted, though it is not an 
imprinting in Dr. Lorenz' sense; it can be shown to depend on the 
same mechanisms of selection and association that operate in ordinary 


Especially if the experience is traumatic. 

grey WALTER: 

Especially if it is traumatic. This seems to me to be a very important 
possible source of error and perhaps also, as so often in these para- 
doxes, a source of enlightenment, because we may have here the bridge 
between the conditioned response with associative learning and the 
type of imprinting which you describe. I should like, as I say, to 
enlarge on the mechanism of this conditioning process as applied to 
children and as appHed to brain physiology, because I think there is a 
big field here to study in an objective fashion. It is our ambition to 
define and describe the instinctive processes in man in terms of quite 
objective physiological mechanisms in the brain. I think that the 
study of these learning processes might be extremely profitable to 
discuss even at this stage. 


Another point, which was raised repeatedly by Dr. Bowlby, is the 
question of affect and what we might call, for the sake of argument, 
'reason'; the experience of a number of events, and the sorting out 
from this experience of what matters. From the physiological stand- 
point, all behaviour is always a mingling of reason and affect. I have 
never seen an animal — a mammal, at any rate — behave without 
physiological signs both of logico-statistical behaviour and of affect. 
One can show this in measurements and I think a division of: 'this is 
reason and this is affect', 'you've forgotten affect', 'you've forgotten 
reason' has no basis in the truth of what happens. I think that 
certainly anyone who neglects one or the other is wrong; and even if 
we think we are neglecting one, we are not doing so. Perhaps the 
psychoanalyst could say why we like to divide things so that we say : 
I feel one thing in my head and another thing in my heart. I think 
this pseudo-anatomy is a clumsy, archaic device for indicating the 
relative proportion of autonomic participation in a behaviour pattern. 
These are arbitrary and dangerous divisions, and I urge that we 
should try to insist upon this recognition of the unity of physiological 


I think that Dr. Odier has put his finger today on one aspect of our 
discussion which is particularly close to our subject-matter when he 
introduced those two polar conceptions of causahsm and finahsm, 
because there it would seem to me that what he and Dr. Bowlby 
said links up with what we heard from Dr. Zazzo and from Mile 
Inhelder, about the psychobiological development of the child con- 
ceived in terms of the socialization phenomenon. I would like to take 
up something which has been said by Dr. Bowlby, that conscience 
has an instinctive root. Obviously conscience is a phenomenon of 
socialization, and I would be in full accord with him if he said that 
conscience had also an instinctive root, because it is questionable 
whether the genesis of its function can really be referred exclusively 
to an instinctual origin. This leads us into the central problem of the 
super-ego. When Freud defined the super-ego for the first time, he 
did not immediately introduce the term super-ego, he spoke of an 
ego-ideal, and later on he changed over more and more to the term 
'super-ego', and 'ego-ideal' gradually disappeared. But many years 
ago Dr. Odier (1926, 1943) wrote a paper in which he suggested that 
conscience had in fact two roots, one which could be described as 
'super-ego', roughly identical with Freud's 'ego-ideal', and the other, 
which he called the 'super-id', with deep instinctive roots, which 
would have been associated with the super-ego of Freud's second 


conception. Now, it would seem to me that Freud's second super-ego, 
Dr. Odier's super-id, would be what Dr. Bowlby had in mind when 
he said that conscience had an instinctive root. It seems to me there is 
another factor in what we usually describe as conscience which has a 
finalistic structure, and this corresponds to Freud's ego-ideal and 
Dr. Odier's super-ego. In fact, I believe that this difference has rami- 
fications even into such finer psychological points as the moment 
when the super-ego is born. Is it true that the super-ego is purely the 
heir of the Oedipus complex, as Freud supposed, or is it developed at 
a much earlier stage, as Melanie Klein says, or are there two forces in 
the determination of behaviour, one which is causal, and tied to 
instinct, and the other finalistic and defined as motive ? I would very 
much like to hear what Dr. Lorenz has to say about this: whether 
there might be a difference between man and animals, the animal 
perhaps being finally directed in a biological sense, but not in the 
sense of a conscious finality. 


Concerning Dr. Krapf's remarks, I should have hked to speak 
of the super-ego and the structure of the super-ego mechanisms which 
are manifested in the very frequent symptoms of what is called auto- 
punishment. In his famous Memoirs of 1923 Freud exposed his con- 
cept of the super-ego, but he did not differentiate between various 
expressions: the super-ego such as we find it in neuroses, the ego- 
ideal, the conscious moral conscience, if I may say so — because it 
would be frightful nonsense to speak of unconscious moral con- 
science — and finally the ideal of the ego. It is on this subject that I 
published the study (Odier, 1943) alluded to by Dr. Krapf to show 
that in fact four different things were concerned. A distinction must 
be made between what I call pseudo-morality of neurosis and a new 
form of morality linked with motivations from moral conscience 
which itself is linked with what I have called choice or adoption of 
norms because our moral life always has a normative character 
whatever the value or the nature of the norms we adopt. 


This can be envisaged as a socializing process, can it not ? 


Yes. We must evaluate certain aims which determine and stimulate 
our work and our interests, etc. Now this process of evaluating aims 
is very little understood. To my knowledge no work on the subject 

13 193 

exists, although it is at the centre of social life and the evolution of the 
human being. 

Another point on which I should like to give some details is that 
of the difference which I thought I had noted between our results 
and the data of the ethological school. What do we find in animals ? 
A lack of selectivity. They can follow anybody as long as the im- 
printing occurs sufficiently early. Among infants, on the other hand, 
there is an excess of selectivity. The fact that many children have a 
mother-fixation is determined by a factor which has certainly escaped 
the physiologists and ethologists and which nevertheless plays a fairly 
big part in human life. This factor is the need for security which is 
conditioned by the need for love. What, for example, do you think 
of the concept which has been called the instinct of security ? I think 
that this raises a big problem. In my opinion the hypothesis that such 
an instinct exists is useless because I think that the need for security 
is the consequence of a series of experiences in the child. In certain 
states the child experiences very agreeable feelings and in other 
situations, on the contrary, he feels anxiety. Hence his need of 
re-estabhshing the feehng of security by elective mother-fixation. 

Generalizing, then, from this point of view, one might say that the 
infant experiences two kinds of feelings which it is difficult to find in 
the animal: on the one hand fear of his instincts, which is why he 
represses them so energetically, and notably fear of his sadism (the 
English school now postulates the congenital nature of sadism, 
children being supposed to be born with a strong sadistic tendency); 
on the other hand, there is the fear of the exterior world, the fear of 
frustrations and menaces. This fear leads him to see in the exterior 
world hostile forces which he attempts to explain to himself by 
magical thought, animism, etc., by all those forms of thought so 
well described by Piaget. Many children, in fact, see in the exterior 
world a baleful influence which constantly menaces their equilibrium 
and their life : they therefore attempt to defend themselves by tighten- 
ing the exclusive link with a specific object — the beneficial protecting 
object — the mother. From this arises the very common conception 
in aff"ective life of 'all or nothing'. And this certainly is an inevitable 
source of deep frustrations and great deceptions. 


If you will pardon my saying so, I think that in the discussion of 
the last few minutes there has been a horrid mix-up regarding 
causaHty and finality and regarding the subjective and objective side 
of behaviour, I will start on finality. I should assert that 'finalism' 
is as much a nonsense as 'causalism'. Let me explain in a parable. I 


am driving along in my old car, with the finality of giving a most 
important lecture. I am enjoying the wonderful finality of the con- 
struction of my car, how beautifully all its constructive details are 
calculated to help humanity to be enfightened by my lecture — it's 
the finalists who are apt to be so assuming, because I am not. Now, 
suddenly my old car says 'hp! hp! hp! hp! hp!' and stops. 

And now I'll find out something fundamental. I'll find that finality 
is not a force. It cannot be used for traction. It is causahty that is 
pushing along my car — and everything else on this planet and all 
others. The fact that my intended lecture is of the utmost importance 
for the salvation of humanity — in other words, the value of the goal — 
does not help me three whoops in hell. The only 'factor' which can 
direct my activities in the right direction is the insight into the 
causality of the defect. It may enable me to screw out the jet of the 
carburettor and blow out the drop of water that caused my motor 
to peter out. 

Let's take another example. The finality 'factor' won't help a man 
in whose appendix a cherry stone has got stuck, but the youngest 
pupil in the surgical clinic can help him, if he has insight into the 
cause of the illness. And that is exactly why medicine, the queen of 
all applied sciences, is forever occupied with the unravelling of the 
causes of all illnesses. It seems to me very futile to quarrel about 
whether the question of causahty or the question of finality is more 
important. Neither is of any importance without the other. The 
investigation of causality would be quite senseless if humanity were 
not striving for goals. And humanity striving for goals would be 
powerless if it had no knowledge of causality, which gives it power 
to change, at its will, sequences of natural events. This relation 
between the final and the causal question must be grasped; it is 
perfectly simple. Yet finalistic behaviour students persist in accusing 
ethologists of being blind to the finality of things, and they do so in 
spite of the fact that we actually begin most investigations with a 
quest for finality. When I see a queer organ in a new animal, I do not 
start with any 'causing' investigation, how the animal got to have that 
organ, but I start with the question. What is this organ for ? And even 
in putting this question, we must not forget that the whole conception 
of finality can only be apphed to processes moving in a certain 
direction, which certain processes in the universe do. We must not 
forget that the finality of adaptation in the organic world is an arrow 
of direction which we have fixed to the process post factum, and we 
don't see the many abortive trials of nature, the whole vast quantity 
of animals and plants that have become extinct. 

Now as to finality in behaviour, I think I will have to explain 
something very commonplace which concerns appetitive behaviour 


and the consummatory act : one of those commonplaces which are so 
incredibly difficult to see. An animal's behaviour is directed; it is 
purposive, and Edward Chase Tolman (1949) has given a very good 
objective definition of purposive behaviour. Purposive behaviour is 
characterized by the fact that the same constant end or goal is 
achieved in the animal by variable adaptive behaviour. Now all 
finahsts,most prominent among them McDougall's (1933) school of 
purposive psychology, had assumed, more or less a priori, that the end 
or goal for which the organism is striving is nothing else than 
finality itself, in other words, the survival value or the biological 
effect of behaviour. For a school of thought which regarded Instinct 
(with a capital I) as a 'directive factor', it was entirely consistent to 
assume this. The great discovery of Wallace Craig was simply 
this: the organism as a subject does not strive for the biological 
effect of its instinctive activities, but only for the discharge of 
these activities for their own sake. It seems utterly commonplace that 
I do not go to luncheon because I purposely want to get fatter, but 
because I hke eating and shall become most dissatisfied if I do not 
get food. Also it seems a rather rude and crude reductio ad absurdum 
to state that the young man does not try to impress the beautiful 
maiden with the direct aim of becoming, as soon as possible, the 
father of a lustily squalling baby, but for very different reasons. I 
think that in judging and describing animal behaviour, we cannot be 
strict enough in keeping apart the survival value of behaviour and the 
subjective end which the animal or man is striving for. We are quite 
justified in saying 'this animal is striving for an end'. If a dog wants 
to kill the rabbits in a rabbit-hutch and first tries to dig a hole under 
the rabbit-hutch and then tries to jump under it, and then finally 
gnaws a hole into the hutch, then that's variation with constant end. 
Well, so much about finahsm and causalism. There ought not to 
be one single biological investigation which doesn't take both view- 
points into account. 


It is interesting that Freud made this distinction in his 1915 paper 
on instinct (Freud, 1915). And this notion is, I think, crucial to the 
whole Freudian theory of instinct — that the aim of the instinct has 
nothing directly to do with survival. 


I was much impressed that Dr. Lorenz' wording is almost the same 
as Freud's; instinct has the aim of 'abolishing the condition of stimu- 
lation in the source of the instinct'. 



Dr. Odier said that the selectivity of the human child in regard to 
its mother should be so much greater than the selectivity of an animal 
directed by I.R.M.s. Now I am going to ask him how does he know? 
Has he ever tried to offer a simphfied or otherwise different substitute 
mother to a newborn child without previous experience? And, of 
course, nobody would dare to make the experiment, to try a chim- 
panzee mother and let her foster a human child. And I can tell you 
what would happen : she would drop it very soon. I am not prepared 
to accept the statement that the human child's reactions are so 
selective. On the contrary, it is our general experience that the 
I.R.M.s of the higher animals are much less selective than those of 
the lower animals, because more scope is left for learning. 


It seems to me that both the statements are true because the Spitz 
smiling response is very unselective. It is not a smiling at any parti- 
cular person : it is the smiling at an oval object with two black dots 
on it which is moving. But within a comparatively short time — I 
think about a month — the response is given only to the mother. 


The greylag goose gets imprinted to man within ten hours, and it 
will know its mother within about forty-eight hours, and not respond 
to any other goose. It is very curious for the narrowness and speciaU- 
zation of its Gestalt perception that the little goose finds it much 
easier to learn the difference between the individual facial expressions 
of two geese, than between two human beings. One of my students, 
Margret Zimmer, and I each reared flocks of geese this year. We 
consented to 'peck' at each other's children, and even with this 
reinforcement it took the goslings about three weeks to differentiate 
clearly, to know the two of us apart. Miss Zimmer is rather a petite 
blonde girl of twenty-four and, well, you know me ! Yet it took three 
weeks until her geese wouldn't follow me any more and mine wouldn't 
follow her, while you never see any attempt of a little goose to follow 
another goose except its mother after two or at the utmost three days. 
As to Dr. Odier's question of 'I'instinct de securite', my answer must 
refer to the difference between McDougall and the purposive psycho- 
logists, and us. The purposive psychologists who were prepared to 
answer the question 'why' with 'so that' had naturally no compunc- 
tion at all about giving instincts names according to their ends and 
creating as many instincts as there were ends, whereas our concepts 


of instinct are based on physiological mechanisms. As to the instinct 
of security — well, the goal of security may be reached by any number 
of neural mechanisms which all contribute to that end. 

I want to add one point to a question of Dr. Bowlby's which 
interested me very much because it is our chief object of investigation 
just at present: that is the overlapping and interacting of two 
simultaneous instincts. We thought formerly that instincts in animals 
were mutually exclusive. Julian Huxley (1914), speaking in parables, 
said that man resembled a ship with many captains on the bridge 
fighting all the time between themselves, while an animal was a ship 
which was also governed by a number of captains, but when one 
captain popped up in command on the bridge, the others had to 
go below. I have quoted that again and again because I beheved it 
myself. But it is not true in the least. One excuse for thinking it was 
our choice of examples. We were quite justified in seeking the most 
simple examples, just as Mendel searched for the monohybrid type, 
but in reality we find that these mutually exclusive, unmixed instincts 
are about as rare as hybrids heterozygous only in one gene. Every- 
thing can overlap with everything, and the result is very often a quite 
crude superposition of movement just as two automatisms can be 
superimposed on each other in one muscle contraction, so you can 
have two instinctive movements superimposed. In my film of the 
courtship of the mallard duck, you will see a very beautiful example 
of superimposed instinctive activities, A female mallard is trying, 
simultaneously, to execute the movements of inviting her male to 
copulation and inciting him to drive away another drake who comes 
disturbingly near. At first you see her doing the two movements — 
both are movements of head and neck — alternatively, and then the 
two rhythms begin to attract each other by what Holst (1936) calls 
'Magnet-Effekt', until both movements fall into step and form the 
queerest superposition. 

Our assumption is that superposition is the most primitive form of 
interaction between two instincts, and that mutual inhibition is only 
due to a secondary mechanism which evolved later. It is, of course, 
very necessary that the instinct of escaping blocks all others, or that 
sexual instinct prevents fighting or killing the mate, etc. It also may 
be regarded as an argument for our assumption that in pathological 
cases, in which the inhibiting and dividing superstructure gets defec- 
tive, there occurs a mixture and superposition of instinctive move- 
ments which normally never occur simultaneously. If you get down 
to the epileptic fits, where all superstructure is suddenly struck away, 
you get a cacophony of pretty well all the endogenous movements 
that the human has — escape activities, sucking activities, etc. But 
your description of that woman who hated, Dr. Bowlby, and at the 


same time was affectionate to you, strongly reminded me of what some 
fish will do, which primarily are aggressive, but in which the aggres- 
sion has to be inhibited in order to let the pair take care of the brood 
together; this inhibition mechanism is very fragile and very easily 
broken down, and you suddenly get an awful fight between the 



The Cross-Cultural Approach to Child 
Development Problems 

Films presented by Dr. Lorenz and Dr. Mead* 


I want to emphasize two or three things on method first. The 
approach I will be talking from is based entirely on the study of a 
whole community whose place in a series of communities consti- 
tuting a closed culture is known. There may be seven or there may be 
seventy villages, but before one selects a village one knows its size, its 
composition, its relationship generally to the whole of the pattern. I 
make the choice of a village that is itself as closed as it is possible to 
get (which depends on the society), that is clearly demarcated and 
contains a functioning whole ranging from grand-parents to babies. 
I don't study any viUage of more than 500 persons, and I prefer a 
smaller one. I study the whole community so that I know the posi- 
tions and relationships of everybody in it. If I concentrate then on 
particular children or particular families, I know their place in the 
whole structure. If I concentrate on particular pieces of behaviour, 
such as that of the pre-school child or of infants, it is against the 
picture of the whole. It is as if you threw a light on the whole tribe 
and then a brighter light on a smaller group, and a brighter light still 
on a still srnaUer group. For instance, I have a thousand photos and 
ten thousand feet of film of one of the babies in a Bahnese village ; 

* Films presented by Dr. Mead: Bathing Babies in Three Cultures, a compari- 
son of the interplay between mother and child in three different settings — bathing 
in the Sepik River in New Guinea, in a modern American bathroom, and in a 
mountain village of Bali in Indonesia. Part of Character Formation in Different 
Cultures, a series of films produced by Gregory Bateson and Margaret Mead, on 
the basis of field work in Bali and New Guinea, showing the relationships betvr'een 
the way in which infants and children learn their culture and other aspects of 
behaviour, particularly dance, trance, and other forms of dramatic behaviour 
(New York University Film Library). 


of many other babies I have perhaps only fifty photographs. By 
using photographs and film as well as notes, we preserve all the things 
that we don't know enough about to look at, and all the things that 
are going on simultaneously so that one could not possibly write 
them all down. It is possible with these new techniques to carry 
many more unanalysed variables than ever we could before. 

It seems to me that we are working here in this group with two 
problems. The first: trying to define the nature of maturation for 
human beings. What are the minimum points below which it is not 
possible to learn certain things ? What is the whole pattern of chrono- 
logical maturation and the degree of individual differences? Do these 
individual differences fall within a normal curve of distribution, or 
are there extreme temperamental discontinuities that are innate ? My 
expectation is that in the end we shall decide that there are a large 
number of innate individual differences in every society; we have no 
reliable method at the moment of measuring any sort of constitu- 
tional differences that are cross-cultural. My impression after working 
intensively in seven primitive societies is that the range of tempera- 
ment in them is about the same as it is in our own society. The range 
of intelhgence is usually narrower, but that is very largely a question 
of the small size of the group, and the fact that most of the defectives 
do not live, so that you have less at the lower liiTiit ; and your chances 
of a genius are enormously reduced when you are working with a 
people of whom there may be fifteen hundred in all. 

Now the other problem we have to deal with is the role of culture 
in patterning the growing individual. When the anthropologist 
works with one of these communities he first makes a cross-sectional 
picture of the adult culture — the finished mature behaviour, the 
institutional patterns, the language. Then he studies the way indivi- 
duals are inducted into the society and learn the culture. I would 
expect that when we get to the point of being able to specify the 
pattern by which the individual learns the culture and the pattern 
which is represented cross-sectionally in the adult you would recog- 
nize them as very highly identical patterns. This would only be so, 
of course, in a static society which is not changing, where the adults 
have been inducted in just the same way as their children are being 
inducted. But in a society such as the latmul of New Guinea, where 
the grandparents have been held by the arm in exactly the same way 
as the baby is held today, so that every member of the society of 
every age has experienced the same infantile situation, you expect the 
old adult to represent the entire growth sequence and growth 
experience, and to represent it to the young individual. 

We can make distinctions between societies where children are 
reared by their grandparents, and societies where they are reared by 


their parents. In societies where they are reared by their parents, it 
looks as though they develop more curiosity and more adven- 
turousness, and they are less conventional. In societies where they are 
reared by grandparents, they learn as babies what it is like to be old, 
and accept it. A third possibihty is where children are reared almost 
entirely by other children, carried around by little girls and boys or 
just by little girls; this has certain characteristics in common with 
rearing by peasant nurses in that it keeps the child much closer to its 
own bodily processes than when the rearing is done by an adult. 

We assume that the cross-cultural differences of mother-child 
behaviour between New Guinea, Bali, and modern America and all 
that they represent in terms of technology, attitude toward hfe, 
posture, gesture, etc., are learned behaviour; that there is nothing in 
the Balinese that prevents them from becoming as awkward as Euro- 
peans. Within one generation of rearing in Europe of Bahnese babies, 
you would expect to lose the lovely weaving of the fingers. I have had 
one experience of an American child in Samoa, where the dancing 
is not as beautiful but is somewhat of the same order. At two he 
could dance beautifully in the Samoan style. He came back to the 
United States at three, and at ten he was as awkward as any American 
child. The whole thing had probably not been eradicated but driven 
underground, and when I spoke Samoan to him he sniggered as if I 
were saying something pornographic. He has an IQ of 160 but a 
deficiency in written English, and when I analysed his English it had 
Samoan grammatical locutions underlying it. This rather dramatizes 
the points that children from quite different environments can learn 
these things as infants and also that these things are all learned. The 
Balinese hand is hke a monkey's hand. In America, when I stand in a 
line and shake hands with college students, about one in two hundred 
has a hand that is like a Balinese hand, and if I ask the girl what she 
hopes to do, most probably she answers, T am going to be a dancer'. 

On the other hand, if we look at the often exceedingly elaborate 
ritual styles, we find that each culture has picked out the innate 
potentialities of part of the human race and, because of our great 
capacity to learn, has devised ways of teaching the other members of 
the culture to do that particular thing. There is a good deal of 
evidence (of the sort that underlies the writings of Jungian students) 
that there are a limited number of these characteristics that recur 
sufficiently often so that we will have repeats of psychological 
patterns in many different parts of the world where we cannot 
prove any historical connexion. Myths (which as nearly as we can 
tell are quite independently evolved by different peoples) are an 
example. They cannot simply be referred to childhood experience, 
because some peoples don't have them. One people will develop 


them and another will not. There is a possibility of recurrence in 
every society of types sufficiently alike to account for these similarities, 
and then each society takes one or more for its model. For example, 
sometimes a society will pick one temperament for a man and another 
for a woman. Another society will pick the male and female of the 
same temperament, and another society will distort one sex greatly 
in order to bring them into accord not only with the temperament, 
but with the sex of another group. In very complex societies, such as 
modern England or France or the U.S.A., we have, of course, class 
typing by learning, and we have occupational typing. We have our 
Bohemias and our Greenwich Villages, which are socially selected 
mechanisms for people who find one way of life more acceptable than 
another. In primitive societies the range is very narrow, and every- 
body who is going to stay ahve has to be fitted into a tighter pattern. 
Every human culture has to be learnable and usable by everybody 
who is going to survive in that society. You can have a society that 
keeps only half the babies alive; there are very many who do. You 
can have a society that says that people who are born deaf aren't 
going to talk, and they are put in an outcast position. But if you take 
the position that is developing in the Western world, that the people 
who are born deaf are going to talk, you may then have to try to 
modify your linguistic usages so that people who are deaf can talk. 
In a society which has a language that makes lip reading difficult, 
and which also has a large number of deaf people, one would expect 
that in time the language would be modified. If you insisted on 
Braille and had a large number of blind people, then you would have 
to work out a machine that would recognize the letters, or you would 
have to alter your script to make it more adaptable to Braille. This is 
just an extreme example of the point that everything that is learned 
has to be learnable by every ordinary human being in the society, in 
addition to being exceptionally congenial to some temperaments. If 
we all sat here and thought what kind of language we would make up 
if we wanted to make one up, our ideas would differ, but if we wanted 
to use the language, it would have to be modified so as to be learnable 
to every person here. In a sense every culture is a model of every 
other culture and contains all these possibilities, which is why an 
adult can move from one country to another and continue to operate. 
The place where we can conventionalize that is language. Man has 
known for thousands of years that he could learn the language of 
another group. We haven't conventionalized the rest of our cultures 
to the same degree, so many people will go into another society and 
learn nothing but the language (not even the postures and gestures 
that go with it). But if we take language as a model, it seems that any 
human being who has thoroughly learned his own culture and learned 


that cultures can be learned, can learn another as an adult. And that 
learning is of a totally different character from the learning as a child. 


What about the sub-cultures of occupational groups ? 


Well, they can be learned if you learn that you can learn them. 
People in the natural sciences are generally taught that other people 
can't learn them. The scientists say: 'They will never understand; 
they can't count.' In a class society, the members of the lower class 
are taught that they can never learn to be members of the upper class, 
and if they are taught that well enough the class society will last for a 
very long time. So when one says that every culture is learnable by 
every member of every other culture, the statement contains all of 
this 'learning to learn' in it. Hebb (1949) discusses in his Organization 
of Behavior the probability that first learning is of a very different 
order from adult learning. Even if so, however, each culture contains 
the record of that first learning in its adult institutions and patterns. 

I want to say something about the application of some of Gesell's 
work cross-culturally. The book Growth and Culture that Frances 
Macgregor and I collaborated on was an ex post facto attempt to do 
this (Mead and Macgregor, 1951). I had not been very much 
interested in Gesell's work until in 1945 I saw what Frances Ilg was 
doing with it, with her enormous capacity to assess kinaesthetic and 
motor development. We worked with four thousand photographs of 
eight Balinese children on whom we had accurate age records. (And 
they were the only infants on whom, in two years' field work, I had 
been able to get both dates of birth and long sequences of photo- 
graphs — this in a society that keeps accurate dates for the first seven 
months of life! I wish everyone who has comments to make on 
material in primitive societies would bear this in mind — that it took 
two years to get sequences on eight children whose date of birth we 
knew with absolute certainty.) We worked with these photographs 
along with the whole Gesell group, spreading the photographs out in 
various simple categories and letting the Gesell people pick out from 
our children the things that struck them as completely deviant. Then 
we took their categories and studied them. The general finding was 
that the gross chronological stages were identical in the two cultures. 
It has been my general feeling (working always with approximations 
and very few real ages) that the sort of thing that Professor Piaget's 
and Professor Wallon's school are agreeing on in periods from five 
to seven or the period about twelve is reproducible cross-culturally. 


What we found was this sort of difference. The BaHnese child's 
hands and fingers are very dehcate. The lovely type of precision that 
one gets with the Balinese is an earlier development than our type of 
prehension and is accompanied by very poor thumb-forefinger 
opposition. When the child picks things up it opposes its thumb to 
the second and third finger and not to the first and tends to bring the 
fingers down to the thumb rather than bringing the thumb toward 
them. The sort of thing that we think of as using the thumb is under- 
developed, and there is a high development of the ulnar side of the 
hand. Prehension gradually develops, and the children pick things up 
with the third, fourth, and fifth fingers, and there is an emphasis that 
gives a tilt to the handling of things. In the same way, the develop- 
ment from crawling to sitting differs somewhat. American children 
go through a long period of either hitching or crawhng, walking on 
all fours, then they stand, and much later learn to squat. The 
Bahnese child is not permitted to crawl very much and is always 
being picked up because crawling is behaving like an animal. It goes 
from sitting to squatting to standing. The squatting is a much deeper 
and broader posture and as nearly as we can tell is an effect of being 
carried on the hip but not having to grasp because of the sling, so that 
the child just spreads. The mother just hangs the baby up on her; or 
there may be a child-nurse who plays violent running games with a 
baby in a sling around her neck. The child has to learn to adapt to 
this. If it didn't, it would have its neck broken, and conceivably an 
occasional baby does get its neck broken. So, apparently a combina- 
tion of the cultural attitude about animahty which the Balinese are 
apt to stress (one tries to keep one's animality under control) and the 
method of carrying babies produces a posture which is a different 
kind of squat from ours. 


Are there vipers or any other dangers to crawling? 


No, these are urban people. The prohibition is directly related to 
the feeling about animality and incest. People who have committed 
the crimes that are most disapproved of, bestiahty and incest, are 
made to crawl on the ground like pigs, eat from a pig's trough, and 
wear a pig's yoke on their necks, in expiation for their behaviour, 
before they are banished from the high gods and condemned to the 
vicinity of the cemetery for ever. 

I want to stress, for a minute, the interaction between the mother 
and the child that we get in these living societies, that gets left out 


here when one keeps talking about 'the child does this', and 'the 
child does that'. A Balinese baby is limp and soft, but at the same 
time capable of performing quite elaborate acts even so young. The 
teaching is kinaesthetic (you get behind the child and you teach him 
by manipulating his body) or it is visual— never verbal. The Bahnese 
are incapable of learning anything verbally. They are incapable of 
carrying out the Binet-Simon Grade III instructions. The Gesell 
phrasing of it is that the child keeps the neonatal flexibihty that we 
see in very young babies. You get the same kind of weaving flexi- 
bility; you see it in Spitz's films on regressed, neglected hospitahzed 
babies — the wandering hands with each finger going off on its own. 
The hand postures are exceedingly asymmetrical. In Bah any 
part of the body can go into trance. The Httle finger can go into 
trance and so can the whole hand. There is catalepsy of parts, and 
there are ceremonial trances in which the hand goes into trance 
but the rest of the man doesn't. The assumption is that you have 
here a mother who expects her baby to be hmp, who embodies in 
every movement she makes in her handling of the baby an expecta- 
tion that the baby will continue to be neonatally flexible. The baby, 
then being flexible, again reactivates the mother's expectation of 
flexibility, and the two things interact. If you could give a baby of this 
sort to a Western mother, you would alter her behaviour to some 
extent, especially if she were an experienced mother and could feel 
the surprise in the baby. On the other hand, if you gave an American 
child of nine months to a Balinese mother, it would probably learn a 
good proportion of this behaviour. My suspicion would be that the 
one girl in two hundred that I encounter in a group of American 
students whose hand feels Balinese has some hereditary predisposition 
that is like the hereditary style that the Balinese have made into a 
whole culture that is learnable, and learnable at different stages. 

I want to make one remark about thumb-sucking, because I think 
it fits into what Professor Lorenz was talking about in connexion 
with I.R.M.s. One of the most astonishing results is that we have 
never found thumb-sucking in genuinely primitive babies that haven't 
had any Western-type public-health nurses around. Even in societies 
with enormous oral emphasis where they suck everything else — their 
wrists, their hps, their knees — they never suck their thumbs. That 
does not mean a thumb never goes into the mouth, but a typical 
thumb-sucking position with the thumb deeply in the mouth is 
absent. The only explanation of this that I can suggest is the kind of 
suckling the child receives as a very young infant. Just once in my 
experience have I seen a baby put its thumb in its mouth in the thumb- 
sucking position — at about two hours after birth, when there was no 
one there to suckle it. 



There is evidence that thumb-sucking probably takes place in 
utero. Swallowing of the amniotic fluid certainly takes place and the 
thumb is in a very convenient position for getting into the mouth. 
At Caesarian operation the thumb has been observed to be in the 


One sees newborn babies as they come from the delivery room, 
before they get to the nursery, who are sucking their thumbs. 


I am suggesting that thumb-sucking represents some sort of later 
phase of deprivation, that if a baby does not have enough breast- 
sucking at a given period it will take to it again, but properly breast- 
fed babies are babies that are fed a great deal and fed when they are 
hungry and without that long period of wait at birth. The thumb- 
sucking is absent in every group where the baby is fed by somebody 
within an hour of birth. 

I would like to close with one other general point. We have a great 
deal of evidence from clinical material that man is capable of receiving 
exceedingly specific irreversible impressions. The sort of thing that is 
called in technical slang 'a lech' has the type of specificity that is 
extraordinarily suggestive of some phenomenon related to imprinting. 
A 'lech' is the sort of sexual demand that makes a man go to a 
brothel because he wants something exceedingly complicated. He 
must have a woman in a white dress with a red sash with spots of ink 
on it, or something. Usually his demands contain elements which are 
uncongenial to almost all wives. It is the sort of thing that the brothel 
seizes on. 

We also have evidence that human beings are capable of making 
self-selection of foods that are right for their own growth; experi- 
ments have been done on rats that show that a rat, confronted with 
prepared substances which never existed in human history in a pure 
form, can pick from among them a better diet than a scientist can 
choose for the rat's particular needs (Richter et al., 1948). 


But it has been suggested that it is a learning process. The first 
try makes you feel better than the others and therefore you concen- 
trate on this. 



What I meant is that this is a potentiality, whether it is completely 
innate or learnt. 


It needs one thing to be added to that, which rather dramatizes it. 
The rat when pregnant will shift to a diet that is appropriate when the 
infant rat is born, and it shifts again when the weaning takes place. 
And when the adrenal cortices have been removed the rat readjusts 
the diet promptly, taking in enough salt to keep himself aUve from a 
salt solution which is available alongside the fresh water, and which 
he ignored previously. A similar thing is true of the rat made 
diabetic. The rat shows an extraordinary capacity to adapt himself 
to his bodily needs in a way which is very hard to conceive of being 
learnt. Of course, it has survival value and in that sense may make 
him feel better. 


We know that human beings will do some of these things. There 
are cases of children without an adrenal cortex who could not yet 
speak, who have learnt to eat salt. There is a very particular case of 
that, an instance in which an adrenal-deficient child had learnt to 
eat salt while the mother was sick and the mother did not know of 
this. The child was taken to hospital, had no way of finding the salt, 
had no way of saying that it needed the salt, and it died. 

Now the difference between building a diet which will keep most 
people alive comfortably and properly fed, and creating a situation 
within which each person can self-select the exact items which he 
needs, represents one of the dilemmas which society has always dealt 
with. It looks as if it is wiser to create a more generalized pattern 
and have people learn. This can be generalized to the point where 
we do not coerce individual idiosyncrasies, where it is possible, for 
instance, for children to refuse to eat something. In some societies 
children are forced to eat things and some of them probably die 
because of it. American share-cropper children are forced to eat fat ; 
they rebel violently but unsuccessfully and that gives them an inability 
to learn to eat meat later on. Something has evidently been forced 
upon them that is nutritionally so bad that it creates a learning defect. 
In every culture you see a compromise between the finding of a 
pattern within which all individuals can survive (though with different 
degrees of free activity), and the insistence on a pattern that is lethal 
or destructive or deforming to some individuals. 



I would like to ask a question. First, I think that we ought to ask 
whether the special requirement in rats leads to the choice of that 
unknown food factor at the first time, or whether it is found by 
trial and error (which I strongly suspect). I think that the answer is 
enormously important, because we know of cases where there is a 
most specialized I.R.M. present for such a special need, for instance, 
the I.R.M. for chalk in birds. Its optimal object simply consists in 
something hard and white, but just not too hard to be nibbled. Any- 
thing which fits that I.R.M. would be eaten by birds in need of chalk 
and I have killed valuable birds by giving them calcium carbide, 
which I mistook for chalk. It just shows that there may be an I.R.M. 
which is never used in the wild bird. 

Now I can go on with what I have to say. Different people say that 
there was a tremendous resemblance between Dr. Mead's film on 
babies and mine on birds. Now I want to give a warning; you saw a 
similarity but you did not see the same thing. You saw something 
analogous, but on a very different level. In both films you saw be- 
haviour patterns which have evolved historically. But in one case they 
evolved in the palaeogenetic history and in the other in the cultural 
history. Both ways of evolving differ from history in the common 
sense of the word in being much slower and much more conservative. 
This forms a link between the two, and it is this conservatism in 
how you court, or how you hold your child's hand, or how you dance, 
and so on, that is responsible for the tremendous similarity, because 
it produces ritualization. In my opinion, the analogy between ritual 
dances and ritualized instinctive activities lies in two points: first, 
activities which primarily had mechanical function, like sowing, or 
reaping food, etc., have been in turn made into something with an 
entirely different symbolic function. Second, several independent 
movements, joined on to each other by an adaptation to the plastic 
needs of the situation, have been welded together into one rigid form, 
which, by that very process of welding, achieve autonomy, as one 
pattern of behaviour. 


I agree with the welding point, but I would question the slowness 
of the process. In an unpublished study on hand postures in court- 
ship (in the biological sense rather than the social), among young 
people in Louisville, Kentucky, Ray Birdwhistell distinguished 
over twelve ritualized pieces of behaviour in the way the average 
college boy in Louisville holds a girl's hand. They are stylized and 
ritualized and not articulate. 




Are they not a fashion ? 


No, they are not a fashion in the sense of being articulately com- 
municated; but I don't have any reason to believe that they are a 
hundred years old. They are systematically linked with so many other 
things in the human body; the body as a vi'hole is a factor in the 
ritualization. In New Guinea we have no indication at all that some 
of the cultures are more than five or six generations old. They change 
with extraordinary rapidity and change into very different styles, 
with the human body as the integrator. 


Gesture is a convention; it has to be learned. There are innate 
elements in it, though. Your directing your hand towards me so that 
everybody understands your meaning is a convention; but it evolved 
historically from a series of innate movements. 


But you could not call the stiffening of the arm in anger a conven- 
tion. It is a conventionalized use of something which originally arose 
in another situation. 


The New Guinea temper tantrum gesture, you mean? It is abso- 
lutely characteristic of children in that part — it occurs over and over 
again. You can say that babies learn to do that because they see 
other babies doing it. You can say they do it because of the way the 
mother holds the arm. It can be learnt very fast though. That is an 
important point which I think we should go into here, because in 
cultures like modern America the speed of change is so great that 
there is very little ritualization. 


I want to ask if anything is known about the extent to which the 
ritualized gestures, either in birds or lower mammals, or in human 
culture, could be related to some complex of physiological mecha- 
nisms. These are things about which we know very little, but which 
are described in great detail in some of the Eastern cultures, particu- 
larly in yoga, where you get highly ritualized postures of different 


types which have a most remarkable effect on the autonomic system. 
In the Indian physiological textbooks of a few centuries ago one finds 
all we know about the vagus nerve, for example, described in terms 
of kundahni, a great snake in the body; and very precise descriptions 
are given of how to control this section of the autonomic system, 
by tucking the heels up under the genitalia and so forth. These look 
like merely ritual gestures, but in fact are very crafty adaptations of 
the human frame for a certain cultural need of yoga or trance. 


I wouldn't say they were ritualized; they are dictated by the needs 
of the body. 


Is it possible in the human that imprinting, if it occurs, is of a 
physiologically fundamental kind so overlaid with embroidery that 
we practically never see the pure form of the imprinting ? 


I hardly know how to answer — of course, imprinting is always 
dependent on some I.R.M. being there, and Dr. Mead was quite 
right to say that there is a tremendous variation in this respect. That 
unhappily deprives you of the only possible character by which you 
could recognize imprinting and the I.R.M. at all without doing 
experiments. I agree with what Dr. Mead has already said of the 
innate components of human behaviour. They must be something 
occurring in very many cultures — that's about all you can say. As to 
experiments, I must ask you not to expect too much knowledge about 
imprinting in man from ethologists. There is one I.R.M. definitely 
found in man, that is the I.R.M. to snakes. This I.R.M. is present in 
about 50 per cent of people and in very many people it is already less 
selective than in some others. It might interest you to know that the 
horror of mice is due to the same I.R.M. as for the snake, because of 
the movement without legs. Snake movement has the legless character 
and also a winding motion, and there are people who do not mind 
the winding but do mind the legless movement, and those are the 
women who are scared of a mouse. 


Could you quantitate the extent of that I.R.M. in an experimental 



Well, in human beings it appears you cannot quantitate it because 
you cannot quantitate the inhibitions of the higher central nervous 


If you could it seems to me that the genetics of it could be studied. 


It begins to operate at about three years — in the three-year-old 
child you can nearly neglect the effect of self-control. 


You have to have people who live in an environment where they 
know nothing about snakes, there are no words for snakes, no dances 
about snakes. 


These were children in Viennese hospitals who certainly lived in a 
world where they weren't taught anything about snakes. 


Were there mice? 

There were no mice. 


One can distinguish between, and to a certain extent compare, 
different cultural ritualisms and the codification of mimicry. One 
example which seems fairly clear to me is that of the different stages 
of smiUng: the smile of appeasement after food which appears 
immediately in the infant and which gradually takes on significance 
and becomes a sign, an expression. It is fairly clear that we have here 
a tonic, postural origin of certain signs. Now, I should like to know 
to what extent this explanation, which is valid for smiling and certain 
kinds of behaviour and some mimicry, can be validated for the 
organization of certain cultural rites. 



I am not quite sure that I understand your question. You have 
people who reply to the smile, mothers who spend hours smiling at 
their babies and playing with them. In other cases, the baby spends 
almost the entire time held on the mother's back in a bag, where it 
has all the cutaneous pleasure of being next to her. When it is hungry 
it can be fed, but it is much less face-to-face with the mother. The 
Balinese regard indiscriminate smiling as the first sign of insanity, and 
one of their definitions of an insane person is somebody who will 
smile at anybody and who will respond to a smile with a smile. The 
only adults who smile are the type we would probably classify as 
hebephrenic. The Balinese are the most unresponsive people in the 
world. They ignore oratory. If you try oratory on them they either 
go home or go to sleep, but you cannot make a speech to the Balinese 
and be heard. Probably this comes from their continuous stimulation 
of little babies. They smile at them and play with them, and they 
don't discriminate between smiling and tears in Bali. It is just as 
much fun to make a baby cry as to make it smile. They just want to 
make it do something. At first there is an exaggeration of the elicita- 
tion of smiHng responses, until the baby is seven months old. The 
Balinese say 'as happy as a baby at its seven-months' birthday', and 
it is just at the height of its smilingness, responsiveness, gaiety. But 
by the time the child is about two-and-a-half, it has decided to quit. 
It ceases to respond. The adult finally does not respond when other 
people want him to do so. The Balinese smiles when he wants to 
smile, not when other people want him to, and grief is not permitted 
at all except on the stage, or in a mother of a baby dying under three 
months. If she cries a Httle then, they forgive her. Otherwise, no 
crying, no grief at all, though on the stage the representation of grief 
would be recognized by a European as beautiful deep grief. 


So you would see an expression of grief, perfectly intelligible to 
us, on the stage but not in real life ? 


Yes, on the stage you mourn when you are lonely, when you are 
deserted, when somebody dies, and it is a beautiful thing. 



I would like to point out that in thumb-sucking, and in relation 
to psychoanalysis and the neuroses something very definite and 


complicated comes in, which has not yet been discussed at all, and 
that is displacement activities. 


Would you give us a summary of what you mean by displacement 
activity ? 


In certain situations, especially those of a conflict, or in other cases 
in which an activated drive is deprived of its normal outlet, there 
occurs a very striking phenomenon, called by Tinbergen (1951) dis- 
placement activity. If an instinctive activity is released, activated by 
its normal I.R.M., and then another, conflicting, instinct is brought 
into play, which prevents the consummatory act from occurring, 
there happens something very surprising: the animal suddenly 
performs movements pertaining to neither of the two conflicting 
instincts, but to an entirely independent third one ! When the fighting 
drive is activated in a stickleback and, at the same time, escape 
reactions are released, because the fish, though furious, is considerably 
afraid of his adversary, he will, after some vacillation between attack 
and retreat, suddenly start to dig in the sand, performing exactly the 
same instinctive movements as when building a nest. Two roosters, 
in the same conflict situation, begin to peck at the ground. Avocets 
do some displacement sleeping (the most amazing among aU displace- 
ment activities), putting their bill behind the wing and, in this peaceful 
position, glaring furiously at each other with one eye. As you afl 
know, humans, in conflict situations, start to scratch their head. In aU 
displacements, it is usually a very common, primitive instinctive 
action which is released. Most frequent are so-called 'comfort 
activities', preening, scratching, yawning, etc. ; perhaps you know the 
fits of displacement yawning in furious monkeys, especially baboons. 
Sucking certainly also is, among all mammals, a very archaic instinc- 
tive movement and it has been known to appear in displacement 
situations in apes, monkeys, and bears. I have no experience with 
thumb-sucking in children, but I am ready to take a bet that you 
win find it is a true displacement activity in moments of stress. In 
man and very highly organized mammals learned activities may also 
occur as displacement : women patting their hair, men adjusting their 
tie, both lighting cigarettes. Always these learned movements are 
highly automatized, kinaesthetically 'ground in'. 

The term 'displacement activity' is, in my opinion, not a good one. 
It was chosen by people who did not know or care about what it 
means in psychoanalysis. 



I want to emphasize that terminological difference; 'displacement 
activity' is where the tension belonging to one instinct is discharged 
by being short-circuited to another instinct activity. If we use this 
for the ethological concept, the term 'displacement' can continue in 
psychoanalytic usage to mean not hitting 'x' but hitting 'y' instead. 
In displacement it is the same instinct being discharged, but with a 
different object (see Tinbergen, 1952). 


The first man to describe displacement activity was Makkink 
(1936). He talked about sparking-over activities. In all conferences 
about terminology I still advocate the term 'sparking-over activities' 
because that's exactly what they are. 

I think much emphasis should be placed on the fact that there is 
first activation and then frustration. There is a little lack of clarity in 
Tinbergen's terminology because with him the word 'block' is used 
for the inhibition sitting on the automatism and removed by the 
I.R.M. But in the case of displacement activities he talks of a drive 
which is 'blocked', when it really ought to be 'reblocked'. An instinct 
must first be 'deblocked' and then 'reblocked' by a conflicting one. 


If you have a person who when frustrated in love suddenly begins 
to play chess very hard, or a woman who after a love disappointment 
takes to eating on a large scale, would that be comparable to displace- 
ment activity? 


On a higher level, it may be, because if you ask me what analogy I 
see in psychoanalysis to displac,ement activities, I should say, I see 
it in two cases : in the neurotic symptom, and also, or at least some- 
thing very like it, in subhmation. 


Do such things as displacement activity and imprinting in fact 
occur in mammals, and if so, under what circumstances ? 


Well, to the question of mammals I have got a very simple answer. 
We don't know a thing about them, because of the purely technical 


difficulty of rearing mammals from immediately after birth. We have 
tried to, we are working on it, but while for birds there is a tremendous 
tradition since the time of old Pastor Brehm, who taught us how to 
rear birds from the egg, there is no tradition on how to rear mammals. 
There is hardly even adequate knowledge of the milks of different 
mammals. There is no mammal which I should undertake to rear 
except big ruminants, let's say a sheep at least, and they take up a lot 
of room, and we haven't done it yet. We must keep in mind that an 
animal who is only just living is no use to us for our experiments. 
You remember that shrike who showed a serious diminishing of its 
I.R.M.s, when it did not get the right food. You can imagine what 
happens when you try to rear a dog or a sheep artificially : it would not 
be absolutely healthy. We haven't succeeded in rearing a newly-born 
rodent, except a guinea-pig, at all, and guinea-pigs are domestic 
animals domesticated by the Aztecs and show a tremendous varia- 
tion, so that they are extremely inadequate for investigation of 
I.R.M.s, Then I must still further emphasize the immense difference 
in the sensory apparatus between most current mammals and man — 
a mammal which is so purely optico-acoustical as man is very hard to 
find, except for monkeys. Of course, if I had had chimpanzees, I 
would have done imprinting experiments with them, and then I 
could tell you a lot about the I.R.M.s and the resulting mechanisms. 
But, unless you take primates, which we intend to do when we have 
room and money, I cannot answer your question. Maybe in about 
five years I can just tell you something about small monkeys, or 
lemurs, with which we intend to start. 


What about the 'Mary had a little lamb' story? It is believed 
among English farmers that, if the lamb's mother dies, and the lamb 
is hand-reared, it afterwards becomes a complete pest to human 
beings. It's always about the house, it follows people, and this is the 
origin of the story of 'Mary had a little lamb, that followed her to 
school'. What effect does that have on the relationship of the lamb 
to sheep, both as a lamb and as an adult? 


Once you try to evaluate animal stories, connected with people 
who do not know what is important and what is not important, you 
find that they have absolutely no value. You just can use them as 
hints, and at present there is one lamb known, that was brought up 
not by Mary, but by Hediger, in Basle, that is definitely imprinted 
socially to man. The imprinting has been quite irreversible, but 


unhappily the lamb is a female, and a female sheep is very passive, 
and there is no indication whom she would like as a mate, so there 
is no telling what man or dog or god-knows-what this sheep would 
like to copulate with. 


I think Dr. H. S. Liddell had one lamb that was reared in the home, 
and I believe I am right in saying that this lamb did not run with the 


Nor does that of Hediger. We must keep in mind one aspect of 
social animals. The most disagreeable thing you can imagine for a 
social animal is a closely related species. If you have greylag geese in 
a flock and a flock of bean geese flies by, they just look away. That 
is not innate, but certainly acquired. A new goose, not personally 
known to the others, is a pariah. A quite definitely negative social 
reaction is shown in an animal which belongs to one group towards 
the next group. 


Do you attribute that to continual disappointment through false 
recognition ? 


It might be. I might add something about the false recognition. 
You know what Seitz called 'Gespenstreaktion' in fishes — their 
reaction to ghosts — which we might call, in a less emotional way, 
the reaction of the broken-down Gestalt. If you change one essential 
character of this quahty you change the whole complex and the whole 
Gestalt breaks down, rather as if one important chord or simple tone 
changed in a melody. This is very disagreeable. There are many 
instances where animals react to the change of single characters with 
escape reactions of extreme intensity. For instance, jackdaws are 
very afraid of a white jackdaw, an albino ; so much so that you could 
not get such an intense reaction in a jackdaw from any other object of 
the same size. The same applies to fish presented with a formalin 
preparation of the same species. That's about the worst thing that 
can happen to them. Now, if you imagine that suddenly a man who is 
a formahn preparation walks in at the door, it would be very 
frightening indeed. He would have white eyeballs, a chalk-white skin 
and so on — a ghost, you see. Martina, my pet goose, who was tamed 


to me as mother, and next to me to my wife, gave a fearful escape 
reaction when my wife sat in the canoe in my place. She didn't 
notice at once, and followed the canoe, when my wife pushed off. 
Suddenly the goose looked up, saw my wife, and incontinently 
dived, and came up yards away. That is a reaction of broken-down 
Gestalt, and something hke that may be one of the reasons why the 
bean-goose may be a 'ghost' to a greylag goose. In favour of this 
explanation is also the fact that the animal can very easily get over its 
fright, and get habituated to, for instance, the white jackdaw. 


This is one of the features of animal behaviour that can be very 
beautifully demonstrated on a neurophysiological level. If one pre- 
sents a human subject with a time Gestalt, that is a time pattern which 
is regularly repeated, it produces very complex responses in the so- 
called association regions which usually die away rather quickly, 
after a few seconds or minutes. Now, if you change one feature in the 
Gestalt, so that instead of making your flashes in one regular rhythm 
you change to another, then the whole picture explodes, and you see 
on the screen an electrical explosion as it were, which is far greater 
than that produced by any single item. Some people say that they feel 
'swimmy' or faint or light-headed — they suddenly have this shuddering 
feehng very much as though they had seen a ghost. 


Is this mechanism related to the kind of horror one feels at the 
sight of anencephalic newborns and the rejection of them by the 
nurses who don't want to feed them, and the mother who wants one 
to die? 


Yes, I should certainly say so. But it might be that in that horror 
an I.R.M. also plays a role, because the anencephalics are just 
the opposite of the normal baby with its protruding forehead and 
typical proportions. Then I might remind you that all pictures of 
ghosts, and all devil masks, and even the Chinese dragons, are distor- 
tions of the human form; they are not animals at all, but they are 
men in a reptile skin with the head proportions of a man, a man 
with horns and a tongue. 

I want to ask Dr. Grey Walter : do you find a correlation between 
the degree to which the Gestalt has become familiar and the intensity 
of the reaction to the breaking-down of the Gestalt? 



Yes, for some people, but that raises the enormous problem of 
types of personality. Some people show this, others show the reverse. 


Yes, the Gestalt perception is very different in different people, and 
it is to be expected that very different changes are necessary to 'break 
up' the recognizable quality. But this also depends largely on personal 
experience with the particular Gestalt in question, with its familiarity. 
Some animals, for instance my ravens, resent the slightest change in 
a very familiar environment, and do not mind it in a less familiar 
one. A new pile of wood in our courtyard would frighten them away 
for days, but the same pile on a forest glade two miles off did not 
matter to them at all ; they sat on it. 

Now, I have been asked to talk about innate releasing mechanisms 
in man. I am somewhat embarrassed, and in order to make my state- 
ment tolerably convincing, I must remind you of the difference 
between Gestalt perception and the summatory character of all 
I.R.M.s, of the independence of simple single key-stimuh. The only 
indication we have that something is based on an I.R.M. is that it 
will respond to single key-stimuli, in other words, that it will respond 
to dummies. That the acquired reactions generally do not react to 
dummies is a general rule to which we have hitherto found no excep- 
tion. When we first experimented with a fish, Astatotilapia, we 
thought it to be pretty nearly impossible that the animals should 
possess acquired reactions to the other fish, because then all we knew 
about fish were Tinbergen's beautiful dummy experiments, in which 
he could do anything with both sexes with dummies. When we started 
to work with Astatotilapia, a fish with a very beautifully patterned 
and colourful male, and a cryptically coloured female, we could get 
all the reactions of a male to a dummy male but none of the reactions 
to a dummy female. The fish simply refused to react even to our 
most elaborate imitations of a female. Then my confidence broke 
down and I said: Well, here we have an I.R.M. responding to a 
Gestalt. Only I was lucky to have in Alfred Seitz a pupil beheving 
more than I did myself, and he said 'Well, you must rear them from 
the egg'. We did so. Now, when one of these fish reached adulthood, 
Seitz, who was a very systematic man, said, 'Now we will start with 
the simplest possible model', and he rolled a ball of yellow plasticine, 
and stuck it on a glass rod, and slowly entered the aquarium with it. 
If I had been forced to take a bet then, I should have taken a bet to 
any sum that this fish wouldn't do anything at all, so convinced was 
I of my having been wrong about the I.R.M. and that the isolated 


male would not react to the simple dummy any more than a normal 
male would. But what really happened was that this fish just looked 
at the plasticine ball, spread all its fins, trembled with emotion, and 
started in a frenzy of courting that ball. That was the most dramatic 
experiment I have ever witnessed. And the simplicity of the I.R.M. 
has been demonstrated again and again ever since. 

Was it just the colour of the ball? 


No, it was just an object approaching and not flying away — one 
slowly approaching object. That's all the I.R.M. there is for the female 
in Astatotilapia. 


Did you make beautiful models with all the patterns in them ? 


At first, for the males normally reared, we had made models which 
we could hardly tell ourselves from real females, they were so beauti- 
ful, and we got no reaction whatever. That was with cichhds. And 
cichlids are, of course, much more intelligent than sticklebacks — 
they have a much higher level of Gestalt perception. 

This preamble is all to say that, where we get reactions to dummies, 
we have a very good reason to suspect that there is an LR.M. Now, in 
man, there are quite a number of reactions which definitely will be 
elicited by extremely simple dummies, and among these are all the 
reactions of man to the expression movements of the fellow-members 
of his species. Everybody knows by self-observation that if we use 
animal heads as dummies, and observe people in zoos, we find that 
morphological characters of an animal's head are invariably inter- 
preted as if they were movements or postures of the human head and 
face in expressing some emotion. If the eagle is the symbol of strength 
and courage — actually he is less courageous than a raven — it is all 
because the eagle has, by virtue of the bone-covering of its eyebrows 
and the form and the angle of its mouth, the big eye looking foward, 
something of what we call the 'hero face' of the human fighting male. 
We cannot get away from that, and the emotional value of the eagle 
face is inescapable. Just as little can you escape the stupid super- 
ciliousness of the head of the camel or the llama, and that is only 
because the head is permanently held slightly above the horizontal, 
so that the nostrils are higher than the eye, because the lid of the eye 


comes slightly down, as a protection against the desert sun, and the 
nostrils are narrow slits as a protection against the desert sand. 
These are all morphological characters, nevertheless you cannot 
prevent yourself from interpreting them as human expression move- 
ments and from feeling that the camel is very stupid and supercilious. 
You can go on giving examples for hours and hours, for instance, in 
the mandarin duck, the upward curve from the corner of the eye 
makes you feel that it is smiling, while in the closely allied North 
American wood duck, the female has a white patch under the eye 
and that has got something of a distressed look, as if it had wept. So, 
we all have a lot of these reactions to dummies represented by the 
heads of animals. But you can go down to simplifications which go 
way beyond the simplification of human expression proportions. 
When I was a child a certain type of railway carriage in Austria, 
which had narrow and high windows and had a ventilation slit just 
above the window, had to me a very disagreeable, frightened look 
because, quite understandably, these slits were interpreted as eye- 
brows, and the face was a long face; while the Pullman cars, which 
had broad windows, and a sort of line beneath, looked large and 
happy. Heinz Werner (1933), a Gestalt psychologist, thought that 
this kind of experience, of dynamifying experience of environment, 
was a genuinely primitive character of all experience as such. In my 
opinion, all the dynamic experiencing of environment is due to a 
miscarriage of an I.R.M. directed to the expression of emotion in 
man. We must always remember that one key-stimulus alone in the 
supranormal object may elicit a reaction that is qualitatively the same 
as the whole set of key-stimuli. This simplification may go on so far 
that something Hke the proportions are reacted to as being beautiful, 
even if they are realized in the most exaggerated way. Our own 
reaction to such supranormal proportions is shown by any fashion 

Something interesting about all these I.R.M.s is that all domestica- 
tion characters elicit a negative emotional reaction. When the artist, 
from Greek sculpture onwards, tries to present something ugly, he 
does not represent any old distortion of the human form, but a quite 
definite distortion of being too short and fat and bow-legged. On 
the other hand, the artist can exaggerate the opposite proportions to 
any extent, he can paint a man with shoulders broad and legs 
elongated to any extent, and these exaggerations are taken in without 
any resistance of our sense of beauty. 


I want to be sure that I have got this last point. Do I understand 
that the characters that are used to produce something unpleasant 


will be ones that are characteristic of domestication in animals? 
Short and fat? 


Yes, shortness of legs, bow legs, loss of muscular tonus, sagging 
belly, small and lustreless eyes, etc., etc. In a series of photographs 
representing domestic animals in comparison with their wild ancestors, 
it is quite amazing how extremely ugly the domestic forms appear, 
and they do so to everybody, not only to biologists. 

But now let me proceed to what interests us most, the mother- 
child relationship. One of the best instances of the I.R.M., except for 
the snake, is our reaction to the quality of cute. In literary German 
there is no word for this, but in the Austrian and Bavarian language 
there is the word herzig which implies the verb herzen, to fondle. The 
word which most succinctly means just this and nothing else is the 
American slang word cute. Now, let's look at the properties which 
produce the impression of a thing being cute. The head must have a 
large neurocranium and a considerable recession of the viscero- 
cranium, it must have an eye which is below the middle of the whole 
profile. Beneath the eye there must be a fat cheek. The extremities 
must be short and broad. The consistency of the body ought to be 
that of a half-inflated football, elastic; movements that are rather 
clumsy ehcit the reaction very strongly, and finally the whole thing 
must be small, and must be the miniature of something. Now, if you 
observe yourself and your reactions to different animals, you will 
find that wherever one of these qualities is present, you react with a 
feeling of that particular kind. In the German language, where there 
are many diminutives, this expresses itself in the names given to 
animals. All animals whose German name ends in 'chen' have at 
least these head proportions. If something is a miniature of something 
very big — an elephant, but you must know how big an elephant ought 
to be — then you find how sweet the baby is. You must know how 
long the trunk ought to be in order to interpret the trunk of the baby 
elephant as being much too short for an elephant. I want to stress the 
relational property of those key stimuli. You may even make a 
miniature of a pipe, and you say, 'oh, how cute'. The child is minia- 
ture. Another heterogeneous summation can be demonstrated very 
nicely in animals which, for instance, have not short legs, but long 
legs, as, for instance, foals or lambs. Nevertheless, you find that the 
degree of cuteness is always dependent on the number of key stimuli 
realized in the object. The cheek is very important and the corpus 
adiposum buccae is also very important, and its lack in monkeys 
does much to detract from the cuteness of little monkeys. Also a 
very important part is the tactile stimulus— the rounded behind of 


the baby, because you feel that when you carry it, and monkeys who 
haven't got that are distinctly repulsive. If you see chimps and young 
gorillas side by side (the gorilla is a ground animal and has a large 
gluteus maximus where it ought to), it is infinitely sweeter than the 
chimp who hasn't got it. Now, in order to see whether many people 
have got that I.R.M., we ought to do a mass experiment with 
thousands or miUions of experimental persons. Just this experiment 
has already been done : it has been done by the doll industry, which, 
of course, sells the supranormal object best. The exaggeration of key- 
stimuli can be very nicely shown in the 'cupie' doll, and the 'Kathe 
Kruse Puppe' in German, and if you want facts on what I say, then 
go to Walt Disney's films and see how Walt Disney represents cute 

I want to say one word more on the corpus adiposum buccae. 
Many speculations have been written on its function. Some have said 
that it helps in sucking. Now I don't see why the monkeys, who 
have got a much longer snout, are able to do without it. I propose the 
theory that the corpus adiposum buccae is really a releaser. It is there 
for the very purpose of eliciting an I.R.M. in the adult. That is not so 
speculative as you may think, because quite a lot of the releasers, 
which have definite correlates in the male's I.R.M. to the female, are 
nothing but fat — fat determining body outline, in breast, hips, and 
so on. So that it wouldn't be so very surprising if the corpus adiposum 
buccae should be the same, especially if there is colour on it. 

Now, a dig at psychoanalysts. I agree with McDougall that 
emotion is the subjective side of an instinct, of one particular instinct. 
One qualitatively isolatable emotion is subjective to one instinct. 
Now, I feel that I can in myself very clearly diiferentiate between my 
reactions to the key-stimuh emanating from the female and my I.R.M. 
reactions to the baby. I think that I can, by introspection, assert with 
a quite considerable degree of certainty that what I feel while 
fondling a lion cub, a chow puppy and a baby is qualitatively the 
same, and different from what I feel when sexual I.R.M.s are brought 
into play. I would even go one step further and assert — I would 
perhaps not dare to publish it, but I do dare to say it among friends — 
that I can introspectively recognize the workings of an I.R.M. as 
such. It gives you quite a particular feeling, I should say the ego is 
always surprised at the unexpected and independent reaction of the 
id— I don't know whether I quite succeed in expressing what I 
mean. But anyhow — what I want to emphasize is the value of 
introspection, of psychology in the strictest sense of the word : it can 
tell the objective behaviour student quite a lot about himself, and he 
ought to be interested in that animal, too. 

Now I want to proceed to a rather passionate theme. It concerns 


one of the very few pure instinctive movements of man, a movement 
of expression. We have already spoken about our own reaction to it, 
of our reaction to the 'eagle-' or 'hero-'physiognomy.When you are 
brought into a situation in which the fighting male in you 'believes' 
that it must go in defence of something, then you get, as reaction, an 
expression movement involving the whole human body and one of 
the few real instinctive activities of our species. If you are an enthusiast 
this reaction will occur whenever you are stepping in for an idea, for 
something that ought to be defended, it may be your nation, it may 
be the old school, it may be the freedom of scientific investigation. In 
all these cases you behave in a very singular manner. You feel a 
prickle going down your back, 'ein heiliger Schauer' — and it is quite 
characteristic of the German language that this is holy only in 
German — and I am not making fun at all. Then the tonus of your 
musculature goes up. Your arms are slightly abducted and go forward, 
you make the hero face, and then you are ready to do anything, to 
lose yourself, to forget yourself, in the good and the bad sense of the 
word. You are selfless and ready to die for the society, for the super- 
ego which you are about to defend, and you are also ready to do 
something absolutely foolish. Now, I affirm that a man who has not 
got that back-prickling reaction is an emotional cripple, and I 
wouldn't like to have him for my friend. But a man who has got that 
reaction and doesn't know about it is a danger to humanity, because 
what everyone ought to know about this reaction is this: when a 
chimp is aroused to enthusiasm, that is to say in defence of his 
family, he thrusts forward the jaw, he throws outwards his arms, and 
he fluff's out his pelt. It is very illustrative of the extreme conserva- 
tism of instinctive reaction that in this reaction we not only fluff" out 
a pelt but we still throw out our arms in such a way that the pelt 
will stick out in a direction in which it serves to make our outline 
bigger and more imposing when we are facing or expecting our 
adversary. I think that this dilemma is very characteristic of the whole 
of humanity and of humanity's need to know and govern its own 
instincts. If there are no instincts, if all instincts drop out, then you 
get an emotionless, feelingless model man of the Brave New World. 
If you have this reaction, but don't know about it, then you are the 
victim of any demagogue. 


These matters of the evaluation of and reaction to body images 
have been very much on my mind for some years. There is a vast 
literature about the way in which animal heads and animal expressions 
have been used by the human to signify certain things. It was most 
prominent about the time of the Renaissance, the best-known work 


being by Della Porta (1668) ; this is a very famous series of drawings 
in which different people with different aptitudes were portrayed 
with different animal heads and expressions. This literature also goes 
down to the detail of individual expressive movements, particularly 
in the Indian dance hterature; the various gestures in the Indian 
dance are often referred to by animal names. It seems to me that the 
interpretation Dr. Lorenz put on this may very well be the correct 
one. I say advisedly 'may' because I do not think it is certain that 
these things are I.R.M.s rather than learned. 

Dr. Lorenz mentioned the idea that to make a person's image into 
a devil's image, you change his shape by spreading him out in a 
particular way. I do not believe that you only do that ; I think you use 
caricature. I believe this is a matter of the valuation of a particular 
physique. I have hanging in my office a very amusing cartoon by 
Vicky of two English politicians whom most of you will know, 
Cripps and Bevin, both now dead. It happened at the time this 
cartoon was done that a British ministry put on an exhibition to 
popularize certain ideas, mainly the Marshall Aid Plan for Europe, 
and in this exhibition they had distorting mirrors which made you 
fat and thin. Vicky had drawn Bevin, who was massive and nearly 
spherical, looking at his long-drawn-out thin image in one mirror, 
and Cripps, who was very slender, looking at a totally inflated image 
of himself in the other. Under the first mirror was written 'without 
aid' and under the other 'with aid'. As an illustration of D'Arcy 
Thompson's (1942) famous method of transformed co-ordinates this 
was amusing, but it was also instructive, I think, of a great deal more. 
There are barriers between people which are to do with their different 
physiques but are not due to I.R.M.s — rather to the valuations we 
put upon physique and our identification positively or negatively 
with different builds. People built, for example, hke Bernard Shaw, 
are simply not regarded as human by certain others, and equally 
those of the Bevin build are regarded as sub-human by some of the 
other sort of people. One must be continually aware of the difficulties 
of communication across these barriers, but they may not be diffi- 
culties of an instinctual nature. If they were the devil-distortion 
might be expected to go in one direction only. But I doubt whether 
it does this and I want to ask Dr. Mead what she thinks about that in 
other societies. Are there places where the long, thin ones are despised, 
and made the prototype of the devil ? 


I would be prepared to believe that there are certain sorts of dis- 
tortion in either direction which, if they were too great, would be 




I saw once in the National Geographic Magazine a picture of a 
black chief with a row of about seven wives standing behind him, in 
the order of their favouritism. They were quite exactly in the order 
of their hour-glass form. 


That would be completely reversed in Bali. The Balinese make 
their monstrous figures of the witch out of the most masculine and 
the most feminine characters you could think of The witch has big 
breasts — the Balinese think they are loathsome — and she has hairy 
arms as well. 


The favourites in the picture had little breasts. 


Yes, but there are plenty of societies where that is not true, where 
big breasts are the important thing, and where the tiny breast would 
be rejected. The Balinese reject both extreme male and extreme female 
characters and prefer the male and female that are— well, Geoffrey 
GoRER (1936) once made a crack about it and said you could not tell 
a Balinese male and female apart, even from the front. This is almost 
true. If they are clothed to the waist you cannot tell the male from 
the female, say at about nineteen or twenty years old, from a short 
distance. They devalue curves, breasts of any size or any degree of 
pendulousness, and hairiness or muscles on the male. They are quite 
capable of developing muscles, and if you turn a Balinese into a 
stevedore he develops perfectly good ones. But their way of life is 
such that they have almost no muscular development. So that you 
have to postulate a complex mechanism, part of which might be an 
I.R.M. and part of which is imprinting or learning the type which is 
approved in your own society — whether that type is simply the type 
you see in your parents, highly valued or devalued. 


But don't these Balinese girls have beautiful female sexual propor- 
tions ? 


They are very under-developed as females. They have very small, 
very high breasts. The standard Balinese figure is such that you carry 


the baby high, and the baby drinks from the upturned breast. In 
some cases you get a pendulous breast, but the mother still carries 
her baby up high. To say that that is a beautiful female figure is a 
matter of the aesthetics of any given people. 


Does any people prefer very big pendulous breasts ? 


Yes, there are societies in which the Bahnese woman would not 
get married. 

May I return to the point about domestication; you said that 
characters producing the feeling of unpleasantness are those associ- 
ated with domestication. That seems to me to lay overmuch emphasis 
on biological factors. 


This is, of course, quite true, and I am quite ready to accept that 
this is a simplification. About domestication, it seems to me — and I 
refer to our explicit permission to talk suspicions — that the same 
holds true to a certain extent for the behaviour traits of domestica- 
tion. You hardly find one domestic animal in which sexual behaviour 
and eating behaviour is not increased quantitatively in an enormous 
degree, and the more complicated forms of social behaviour, caring 
for young, and so on, decreased in intensity. This is so general among 
domestic animals that I would find it very hard to find one instance 
of a domestic animal that is more social than the wild one, except, 
of course, the dog. Now, I think that in our emotional evaluation of 
our own instincts we put a very great plus value on those things 
which in domestication tend to atrophy and to disappear, and a 
minus sign to those which tend to hypertrophy. Everything we call 
bestial and brutish is not characteristic of animals in general but 
quite exclusively characteristic of domestic animals. I mean, if some 
wild goat were to talk about the proverbial lecherousness of the 
domestic buck, he would say, 'Poor fellow, he has been under human 
influence for centuries.' I see this daily in greylag geese as compared 
to domestic geese. You get similarities to human beings which are 
surprising, and you cannot help yourself from having sympathy 
with those nice, restrained, non-bestial wild ones and feeling some 
contempt for the domesticated ones. We must keep in mind that 
mother-love is not more necessary to the survival of the species than 
the drive to copulation. Why, then, are those drives to copulation 
"brutish' and why is 'maternal love' sublime? This is simply our 


emotional valuation of instinctive behaviour in man — and it is 
largely dependent on supply and demand. I am convinced that we 
have something very deep, innate, in our behaviour, which tends to 
devalue sex and eating and to value very highly mother-love, social 
behaviour, defence of family, and so on. 


You can find societies which put a high value on sex and eating, 
and a low value on maternity, and it would be important to make a 
Hst of these atrophying and hypertrophying characteristics, and then 
see how they are combined in different societies. For instance, in 
the Marquesas, which is about the most pathological society that we 
know much about, women dislike having children very much and 
maternity, breast-feeding and care of the child are devalued. Women 
are reduced to 50 per cent of the men, so that every woman can have 
at least two husbands, whom she keeps by her sexual attraction, and 
maternity destroys the woman as a sexual object. Food was highly 
valued and cannibalism highly institutionalized. This was one of the 
most vulnerable societies we know anything about; it almost dis- 
appeared with European contact, in spite of a big population and a 
good food supply. Comparison of these points in different societies 
has enormous possibilities if we bear in mind that societies can 
organize these things in different ways. 


Let me ask a very special question of Dr. Mead about the 
Marquesas. We quite agreed on the fact that there is a supranormal 
object and that this supranormal object may cause imprinting in a 
non-desirable direction. We must not forget that in imprinting we 
get all the accidental characters of the object which causes imprinting 
linked up irreversibly with the releasing key-stimuli of the object 
itself. Do you think you can get sexual imprinting of young males to 
an unbiological cover-girl — in your 'pathological' Marquesas sense ? 


Well, Bali is on the edge of it certainly. To have a whole society in 
which the bulk of young males is imprinted to an inaccessible object 
in some way — I do not know of any extreme behaviour like that. Bali 
is a case where you fall in love with your not-mother, or, taking it 
the other way, perhaps, the attitude toward the mother is sexually 
inhibiting to a very marked degree. In all theatricals, the hero tries 
to get a beautiful girl who is not like his mother, and the tragedy is 


that you are always trying to marry the slender, far-away, not-mother 
figure, and you end up married to a Bahnese woman. There is a very 
high degree of avoidance between husband and wife. The marriage 
ceremonies are filled with highly ritualized jokes on the possibility 
that the husband may never consummate his marriage at all and 
never have any children, in a society which is organized to make 
people have children. 

This is not as simple as a simple imprint at all. You have simul- 
taneously presented to the young child the witch, embodying the 
fear of its mother, and the beautiful princess he would like to have 
but whom he can never get. This is represented over and over again 
in a whole series of forms, so that the child is getting treated by his 
mother in a witch-hke fashion and sees the supranormal witch played 
by a very tall man with all the exaggerated characteristics, and at the 
same time sees his other visionary type of sexual object. The Balinese 
define the delights of sex as starting with the first interchange of 
glances and going steadily down, 


Now, just one thing. I want to get your reaction to this. If you 
compare — I hope you will forgive my using a goose and not a mammal 
— a wild goose, you know that the first sign of her love is just a 
glance at him, and when he displays, apparently she looks every- 
where else except at him, but really she does look at him, but so 
quickly that nobody notices it except him: at least, if she doesn't look 
at all, he ceases the display immediately. From these first prelimi- 
naries the whole thing very gradually works up to neck dipping, to 
the triumph ceremony, and so on, until at last she invites copulation. 
Now, a pure wild greylag goose is absolutely unable to come to 
copulation in any other way except beginning with a glance and 
working up through the whole sequence of those reactions. The 
copulation of the domestic goose is totally independent of all these 
preliminaries, and when she wants to copulate she obtrudes herself 
to a male with three or four neck-dipping movements and then 
invites copulation, and that is all. Yet she differs from the wild one 
only in that the activities of lower intensity are skipped over or 
dropped out and the action of highest intensity appears without 
these prehminaries ; the whole step-ladder of intensities leading up 
to it in the wild bird has become unnecessary. Now I think that I 
see this in human beings. There are some that will start with very hot 
glances, and love is likely to be immediate. The activities are the same 
which occur in a 'normal' woman at a very much later stage and the 
difference is only that the preliminaries have dropped out. 


Now the question is, have you got something like these differences 
— of a very long ladder to copulation, or the dropping out — in your 
different cultures ? 


You have whole cultures that represent either type. For instance, 
among American Indians courtship might take four or five years. 
The girl never speaks to the boy, but there would be a shght exchange 
of glances, maybe for two years. Then at corn-husking, if she had 
got to the point where she wants him to do something really rather 
desperate — to speak to her — she might try to get from the ears of 
corn a pure red ear and then carry in her arms a bunch of corn, 
one ear of which was the red one, and as the boy went by he might 
say, 'That is a red ear of corn.' That would do for the next year. 

Then perhaps for another year or so he would go hunting with her 
brother. Then when you finally get marriage you go through a very 
formal presentation, sending game home by her brother to her 
father and so on. These Indians are one of the few people we know 
that have a long period after marriage before copulation, and in 
many of these societies the girls wear a sort of chastity belt for 
quite a long time after marriage. Husband and wife lie awake and talk. 
You have enormous romanticism about the husband talking to his 
young bride — he has been wanting to talk to her for seven years and 
has not had a chance ! This occurs in an area also where we have the 
institution of the copulation blanket, a special leather blanket which 
it is necessary to borrow from a chief, in order to beget a child legally 
and ceremonially, so that it will be the right kind of child, and where 
people boasted of not having children within five years of each other, 
or seven years of each other. It is pattern of extreme restraint and 
extreme romanticism combined. Then you can have societies with all 
the other points in which they behave just like your domestic geese. 
But I see one other even more important thing in what you said, 
and that is that very rapid culture mixture and very rapid social 
change have the same effect on human beings in vulgarizing the re- 
sponses and destroying sequences that domestication has on animals. 


It is always a matter of breaking up or dissociation. 


Of these elaborate patterns: the training of individuals in this 
excessively elaborate series of preliminary requirements before a 
mate could be chosen. 



To come back to what Dr. Lorenz was saying; you could, of course, 
speed up the response considerably by giving a sex hormone. This is 
true of most mammalian species. 


There are two processes working in the same direction. The loss of 
selectivity of the I.R.M.s in the domestic goose is accompanied by a 
tremendous increase in hormones, which is quite clearly indicated by 
the fact that the domestic goose lays eggs in the first year and the 
greylag goose only starts laying properly in the third. On the other 
hand, you get considerable independence of the ceremonies from the 
hormones ; because all these rituals do develop in the domestic goose 
much later in life. She starts by being absolutely promiscuous, 
because she is sexually mature before all her 'marriage' ceremonies 
are mature, and these are what keep the pair together, and make for 
monogamy. So that in domestic geese who do develop a 'triumph 
ceremony' — not all of them do — you have a stage of absolutely 
promiscuous life ending up in the slow development of the triumph 
ceremony, which matures later than normally and in the end there 
results a monogamous couple quite similar to the wild ones. 


In general, I suppose, hormones don't affect the sequence of events, 
but they lower the threshold. 


Yes, they lower the threshold. They may have a big effect and they 
may have no eiffect at all, and that there is no effect on behaviour 
sequence can be even shown by the potential homosexuality of some 
birds. When you haven't got a male, the female may do with another 
female all the things that a male normally does, up to copulating, 
but while she is behaving absolutely like a male, she lays eggs, which 
shows that her female hormones are quite all right. We thought for a 
very long time that in cichlids sexual behaviour was dependent on a 
quite definite hormonal stage which A. A. Allan (1934) thought too 
when he wrote his paper about 'synchronization of the mating cycles'. 
We were quite sure that this was the case with all fish, until this 
year when Beatrice Oehlert found a bigamist fish who had a wife and 
children which had just hatched out of the eggs. He knocked down 
and chased away the male of another couple at the same time in 
another corner of the same tank and occupied his territory and his 


wife. Then he went to and fro and behaved in quite a different manner 
with his new wife and his old wife, doing nest-building and insemi- 
nating activities in one corner, going back and doing the phases of 
three weeks later with his first children, then going back and fanning 
the eggs with his second wife, and so on. This showed that he could 
change his reactions within seconds, quite independently of his 
hormonal state. 



ADRIAN, E. D. (1931). /. Physiol. 11, 132. 

ADRIAN, E. D. and LUDWIG, c. (1938). J. Physiol. 94, 441. 

AJURiAGUERRA, J., ZAZzo, R. and GRANJON, N. (1949). Ence- 
phale, 38, 1. 

ALLAN, A. A. (1934). Auk., 51, 4. 

AMES, L. B. et al. (1952). Child Rorschach responses. Developmental 
trends two to ten years, New York. 

ARMSTRONG, R. K. {\9A1). Bird display, CdimbndgQ. 

BAYLEY, N. (1943). Child Develpm. 14, 47. 

BiRUKOV, G. (1952). Z. vergl. Physiol. 34, 448. 

BOYNTON, B. (1936). Univ. la. Stud. Child Welf. 12, No. 4. 

CONEL, J. L. (1952). In: The biology of mental health and disease, 
London (The report of the twenty-seventh Annual Conference of the 
Milbank Memorial Fund). 

DARWIN, c. (1872). Der Ausdruck der Gemiitsberwegungen bei 
Menschen und Tieren, Stuttgart. 

ELLIS, R. B. w. (1950). Brit. med. J. 1, 85. 

FAIRBAIRN, w. R. D . (1952). Psychoanalytic Studies of the person- 
ality, London. 

FESSARD, A., MONNIN, J., PIERON, H. (1931). Bul. Inst. nat. 

Orient. Prof. 3, 197. 
FREUD, s. (1915). Instincts and their vicissitudes. In: Collected Papers 

IV, London. 
GAMPER, E. (1926). Z.ges. Neurol. Psychiat. 104, 49. 
GESELL, A. (1929). Infancy and human growth. New York. 
GESELL, A. (1941). Wolf child and human child, London. 
GESELL, A. and AMATRUDA, c. s. (1946). Embryology of Bchavior, 

New York. 
GESELL, A. and AMES, L. B . (1945). J. genet. Psychol., 66161, 45. 


(1939). Arch. Neurol. Psychiat., Chicago, 41, 755. 
GOODENOUGH, F. L. (1931). Anger in young children, Minneapolis. 
GORER, G. (1936). Bali and Angkor, London. 
GROHMANN, J. (1939). Z. TierpsychoL, 2, 132. 
HEBB, D. o. (1949). Organization of behavior. New York. 
HEINROTH, o. (1911). Verh. v. intern, ornith. Kongr. Berlin 1910. 
HESS, w. R. and brugger, m. (1943). Helv. physiol. Acta, I, 33. 
HINDE, R. A. (1953). Behaviour, 5, 1. 

HODGE, R. S., WALTER, V. J. and WALTER, W. GREY (1953). Brit. J. 

Delinq. 3, 1. 
HOLST, E. VON (1936). PflUg. Arch. gcs. Physiol. 237, 655. 
HOLST, E. VON (1939). Ergbn. Physiol. 42, 228. 
HUXLEY, J. (1914). Proc. Zool. Soc. Lond. 84, 491. 


INHELDER, B. (1943). Lc Diagnostic du raisonnement chez les debiles 

mentaiix, Neucbatel & Paris. 
INHELDER, B. (1948). Synthese, 7, 58. 
INHELDER, B. (1951). Xllleme Congres international de Psychologie, 

p. 153. 
ITO, p. K. (1942). Hum. Biol. 14, 279. 
KLEIN, M. (1948). Contributions to psychoanalysis, London. 
KOHLER, w. (1940). Intelligenzpriifungen an Menschenaffen, Berlin. 
KOSKAS, R. (1949). Enfance, No. 1, p. 68. 
KRAPF, E. E. (1950). Schweiz. Arch. Neurol. Psychiat. 65, 108. 
KRATZIG, H. (1940). /. Orn., Lpz. 88, 139. 
LE MARQUAND, H. s. and RUSSELL, D. s. (1934). Roy. Berks. Hosp. 

Rep.?,, 11. 
LENNOX, w. G., GIBBS, F. L. and GIBBS, F. A. (1945)./. Hered. 2)6,212). 
LORENZ, K. (in press). Vergleichende Verhaltenslehre, Vienna. 
LUQUET, G. H. (1913). Les Dessins d'un enfant, Paris. 
MAKKINK, G. F. (1936). Ardea, 25, 1. 
MEAD. M. and MACGREGOR, F. c. (1951). Growth and culture. New 

MINKOWSKI, M. (1924). Schweiz. Arch. Neurol. Psychiat., 15, 239. 
MINKOWSKI, M. (1925). Schweiz. Arch. Neurol. Psychiat., 16, 133. 
MONNiER, M. (1946). Schweiz. Arch. Neurol. Psychiat. 56, 233; 57, 325. 
MONNiER, M. and WILLI, H. (1947). Ann. paediat. 169, 289. 
MONNIER, M. and WILLI, H. (1953). Mschr. Psychiat. Neurol, 126, 

239, 259. 
MONNiN, J. (1933). Bul. Inst. nat. Orient. Prof. 5, 1. 
ODiER, c. (1926) Int. Z. Psychoan. 12, 275. 
ODiER, c. (1943). Les Deux Sources consciente et inconsciente de la vie 

morale, Neuchatel. 
PELLER, s. (1940). Growth, 4, 277. 
PIAGET, J . (1923). Le Langage et lapensee chez Venfant, Neuchatel and 

PIAGET. J. (1926). Le Langage et la pensee chez Venfant, Geneva. 
PIAGET, J. (1936). La Naissance de Vintelligence chezVenfant, Geneva. 
PIAGET, J. (1946a). Le Developpement de la notion du temps chez 

Venfant, Paris. 
PIAGET, J. (1946b). Les Notions de mouvement et de vitesse chez 

Venfant, Paris. 
PIAGET, J. (1947). La Psychologie de Vintelligence, Paris. 
PIAGET, J. (1951). Cahiers Int. Social. 10, 34. 
PIAGET, J. (1951a). Traite de logique, Paris. 

PIAGET, J. (1951b). Introduction a Vepistemologie genetique, Paris. 
PIAGET, J. (1952). Essai sur les transformations des operations logiques, 

PIAGET, J. and INHELDER, B. (1941). Lc Devcloppement des quantitcs 

chez Venfant. Conservation et atomisme, Neuchatel and Paris. 
PIAGET, J. and INHELDER, B. (1948). La Representation de Vespace 

chez Venfant, Paris. 


PIAGET, J. and INHELDER, B. (1951). La Genese de la notion du hasard 

chez Venfant, Paris. 
PIAGET, J., INHELDER, B. and SZEMINSKA, A. (1948). La Geometrie 

spontanee chez Venfant, Paris. 
PIAGET, J. and SZEMINSKA, A. (1941). La Genese du nombre chez 

Venfant, Neuchatel and Paris. 
PICHON, E. (1947). Le Developpement psychique de Venfant et de 

Vadolescent, Paris. 
PIERON, H. (1949). La Psychologie differentielle, Paris. 
PORTA, G. B. DELLA (1668). La fisionomie delVuomo, Venice. 
PRECHTL, H. F. R. (1952a). Experientia, 8, 220. 
PRECHTL, H. F. R. (1952b). Natui'wissenschaften, 39, 140. 
PRECHTL, H. F. R. (1953). Behaviour, 5, 32. 
PRECHTL, H. F. R. (in preparation). Quantitative Untersuchungen iiber 

den Greifreflex. 
PRECHTL, H. F. R. and SCHLEIDT, w. M. (1951). Z. vergl. Physiol. 

33, 53. 

PRUDHOMMEAU, M. (1947). Le Dessin de Venfant, Paris. 

RICHTER, c. p., HOLT, L. E. and BAVELARE, B. (1938). Amer. J. 

Physiol. 122, 734. 
SCAMMON, R. E. (1930). In: Harris, J. A. et al. The measurement of 

man, Minnesota. 
SCHOEN, L. and holst, e. von (1950). Z. vergl. Physiol, 32, 552. 
SEiTZ, A. (1940). Z. Tierpsychol. 4, 40; 5, 74. 
SIMMONS, K. (1944). Monogr. Soc. Res. Child Develpm. 9, 1. 
SIMMONS, K. and greulich, w. w. (1943). /. Pediat. 11, 518. 
SIMON, th. (1939). In: Imprimerie Moderne, ed. Centenaire de Th. 

Ribot, Agen, p. 558. 
spitz, r. a. and wolf, k. m. (1946). Genetic Psychology Monogr. 

34, 57. 


CAMPBELL, A. M,, and z YGMUNTO wicz, A. s. (1951). /. cUn. 

Endocrin. 11, 1224. 
TANNER, J. M. (1947). Proc. roy. Soc. Med. 40, 301. 
TANNER, J. M. (1951). Hum. Biol. 13, 93. 
TANNER, J. M. (1952). Amer. J. phys. Anthrop., N.S. 10, 427. 
TANNER, J. M. (1953). Lect. sci. Basis Med. 1, 308. 
THOMPSON, d'a. w. (1942). On Growth and Form, Cambridge. 
TINBERGEN, N. (1951). The Study of instinct, Oxford. 
TINBERGEN, N. (1952). Quart. Rev. Biol. 11, 1. 
TOLMAN, E. c. (1949). Purposive behaviour in animals and man, 

WALLON, H. (1925). U Enfant turbulent, Paris. 
WALLON, H. (1933). Les Origines du caractere chez Venfant, Paris. 
WALLON, H. (1941). U Evolution psychologique, Paris. 
WALLON, H. (1942). De Vacte a la pensee, Paris. 
WALLON, H. (1945). Les Origines de la pensee, Paris. 
WALLON, H. (1946). Egypt. J. Psychol. 1, No. 1. 


WALLON, H. (1947). Cahiers Int. Social . 2, 3. 

WALLON, H. (1949). Les Origines du caractere chez Venfant, Paris 
(2nd editn.). 

WALTER, w. GREY (1950). In: Hill, J. D. N. and Parr, G., Electro- 
encephalography, London. 

WALTER, w. GREY (1953). The living brain, London and New York. 

WERNER, H. (1933). Einfiihrung in die Entwicklungspsychologie, 

ZAZZO, R. (1942). Psychologues et psychologies d'Amerique, Paris. 

ZAZzo, R. (1947). In: Inst. nat. Etude Travail Orient, prof, ed.. 
Etude objective du caractere, Paris, p. 128. 

ZAZZO, R. (1948). Enfance, 1, 29. 

ZAZZO, R. (1950). Enfance, 3, 204. 

ziNGG, R. M. (1941). Wolf-children and feral man, New York. 



Abstraction, and Gestalt percep- 
tion, 99 ff 
Adaptation, physiological, 120-2 
Adolescence, development in, 169- 

time of, 44-5 
Aifect and reason, 192 
Affectivity, 189-90 

fluctuating, 88 
Africans, space interpretation in, 

Age, correlation with developmen- 
tal stages, 88-9 
Allometry, 57 

Alpha rhythms, 140-1, 152-3, 156 
Anencephalics, mesencephalic, 65-6 

meso-rhombencephalic, 64-5 
Anencephalus, rhombencephalic, 63 
Animals, abstraction in, 99-100 
Apraxia, 164 
Athetosis, 73, 74 
Autism, 165-6 
Auto-punishment, 193 
Autorhythmia, 109-10, 118-19 
Autostaxis, 109, 119 

Balance, see Stability 
Bali, child development in, 205-6 
Bearing, reactions of, 164 
Behaviour, environment and 
changed, 51 

purposive, 196 
Biological and Psychological planes, 

relation, 190 
Bowlby, J., biography, 25-7 
Brain, vulnerability of, 136-7 
Brain development, 42, 54 

maturation of parts, order of, 55 

Caricature, 225 

Carothers, J. C, biography, 34-5 

Causality and finality, 194-6 
Circular activity, 163 
Communication, difficulties of, 14 
Conscience, and instinct, 192-3 
Conservation, 76, 77, 79-82 
Constancy computers, 97 ff 
Constitutional differences, 53 
Corpus adiposum biiccae, 222-3 
Cortex, adrenal, development, 43-4 
Critical periods, 47-8 
and allometry, 57-8 
Cultural differences, 201 ff 
Culture, learning, 203^ 
Culture variations, 92-3 
'Cuteness', 222 

Delta activity, 1 34 
Diet, self-selection, 207-8 
Disillusionment, 123, 124-5 
Displacement activity, 214—15 
Distortion, physical, 225-7 
Domestication, 227 

andl.R.M., 113-14 
Ductility, 136, 153, 155 
Dummies, reactions to, 220 

Education, and developmental 

stages, 90-1 
Egocentrism, 180-1 
Electroencephalography, accuracy 
of results, 133 
correspondence with Piaget's 
stages, 149-50 
Emotion, role of, 180 
Enthusiasm, reaction of, 224 
Environment, and behaviour 
change, 51 
and development, 49, 85 
organism's interaction with, 50 
and pubertal age, 57-60 
Epiphyses, union, 46-7 


Equilibration, Piaget's concept, 76 ff 

Ethology, 30 

Exercise, and development, 52-3 

Fading, of reactions to I.R.M.s, 

Failure-to-safety, 137, 148 
Feathers, development, 48-9 
Finality and causality, 194-6 
Flight, maturation curve, 52 
Flight reaction, 73 
Foetus, brain activity, 133 
Form change, 38 
Fremont-Smith, F., biography, 15- 

Function, change of, 98 

Generalization, 69-70 
Gestalt, broken-down, 217 
Gestalt perception, 97 ff 

variations in, 219 
Gesture, ritualization of, 210 
Graphic ability, evolution of, 1 70-2 
Group structures, 76 
Groupement structures, 76 
Growth, and circumscription of 

reactions, 67 
Growth curves, 37^1 
Growth spurts, 36, 38 

localized, 54 

Head turning, ipsiversive and con- 

traversive, 66-7 
Heterogeneous summation, law of, 

Hormone development, 43-4, 231 
Hybridization, and I.R.M., 114 
Hypothalamus, 44 

Imitation, origins of, 72 
Imprinting, 115-16, 130, 184-5, 

Inhelder, B., biography, 30-1 
Innate Releasing Mechanism, 1 10 ff, 
disintegration of, factors leading 

to, 113-14 
in man, 219 ff 

Innateness, 49 

Instinct, dislocation of responses in 
infancy, 184 
Freudian theory, 182ff, 187-8 
overlapping of, 198 
Intelligence, representative, genesis 

of, 78 
Intelligence quotients, and develop- 
mental stages, 91-2 
I.R.M., see Innate Releasing Mech- 
Irreversibility, in reasoning, 78 ff 
of imprinting, 116-17 

Kamala, 95-6 

Krapf, E. E., biography, 24-5 

Language, conventionalization of, 
development of, 172^ 
Lattice operations, 76 
Learnability, 203 
Lorenz, K. Z., biography, 27-29 

Machines, and Gestalt perception, 

Macy Foundation, 13 

Mammals, rearing, 216 

Marquesas, 228 

Maturation, mental and reproduc- 
tive, divergence, 45 

Maturity, sexual, age in boys and 
girls, 57, 58 

Mead, Margaret, biography, 19-20 

MeUn, K. A., biography, 23 

Menarche, time of, 59-61 ; see also 

Mitotic division, 54 

Monnier, M., biography, 16-17 

Motor development, 161 ff 
stages, 164 

Motor functions, integration, 62 ff 

Myelinization, 43, 51 

Myths, 202 

Neglect, neurosis of, 189 
Nervous system, development, 42-3 


Object, development of idea of, 7 1 

formation of, 184 

'good' and 'bad', 184 

specificity of, 188-9 
Object-fixation, and imprinting, 1 17 
Object relations, genesis of, and 

head movements, 70 
Objects, external, movements to- 
wards, 163-4 
Odier, C, biography, 33^ 
Oedipus complex, 189-90 
Ontogeny and phylogeny, in re- 
tarded development, 49 
Operations, mental, S2-4 
Ossification, sequence of, 87 
Overlapping, of instinct, 198 

reciprocal, law of, 170 

Pattern, search for, 140 
Peace, search for, 134, 153 
Petit-mal, 149 

Physiological development, 43-5 
Piaget, J., biography, 31-3 
his conception of development, 

Piaget-Wallon controversy, 165-6, 

Pleasure, search for, 140 
Posture, automatisms, 162 
Prehensile reflex, 119 
Preliminaries, copulatory, omission 

of, 230 
Proportion, teaching of concept, 91 
Psychoanalysis, use of terms in, 1 87 
Puberty, age of, and environment, 

57, 58-9 
course of, 58 

Reactive patterns, change in, 66-7 
Recognition, false, 217 
Reflexes, 108 ff, 117-18 
Reflexive verbal forms, 177 
Remond, A., biography, 23-4 
Responses, instinctual, waxing and 
waning of, 185 
parent -child, 185 
Reversal of behaviour, 67-8 
Reward, 191 

Ritualization, 209-10 
Rorschach patterns, and E.E.G.s 

Sadism, 194 

Scale of measurements, 54-5 

Schema of behaviour, 77 

Scholastic stages, 166-7 

Scyphomedusae, 110 

Security, 194 

Selection of children for study, 1 32 

Selectivity, in animals and infants, 

Self, awareness of, 174-5 

Sinistrogyration, 171 

Size, evaluation of, 105 

Sleep, brain study in, 157 

Slow rhythms, in brain activity, 
133 flf 

Smiling reaction, 72, 184, 189, 197, 

Sociabihty, syncretic, 165 

Spatial relations, Africans and, 103 

StabiUty, 146-7 

Stage of development, criteria of, 
Piaget's concept, 75 ff" 

Stomach, influence in response, 68 

Stretching, 73 

Structures, 75 ff" 
types of, 76 

Struthers, R. R., biography, 17 

Studies, cross-sectional and longi- 
tudinal, 158-60 

Subchoreic instabihty, 164 

Super-ego. 192-3 

Symbolic thought, genesis, 78 

Tarmer, J. M., biography, 17-18 
Teeth, development of, 87 
Temper, and theta activity, 139, 152 
Temper outbursts, 139^0 
Tension, and behaviour, 188 
Tests, aff"ective and intellectual 

factors in, 106-7 
Theta rhythms, 137^0, 144, 152, 

Thumb-sucking, 206-7 
Tonic activity, 162 


Tonus, regulation of, 164 

Typing, class and occupational, 203 

Ugliness, characteristics of, 221-2 

Velocity curves, 38-41 

Versatility, 144-6 

Vulgarization, of reaction to I.R.M. 


Wallon-Piaget controversy, 165-6, 

Walter, W. Grey, biography, 20-2 
Wave pattern, in growth, 46-7 
Wolf-children, 95-6 
Working idiots, 155 

Zazzo, R., biography, 22-3 








OCT 15 'W 

H&L ^ 


3 12 62^5555 0163