Moving towards the Other
The relevance of Hermann Rorschach’s method for the philosophy of perception

Helge Malmgren, MD, PhD
Department of Philosophy, Göteborg University, Sweden

Poster presentation at ”Tucson III”:
Toward a Science of Consciousness
Tucson, AZ, April 27-May 2, 1998
© 1998, 2002 Helge Malmgren, Linnea Malmgren

Hermann Rorschach and the kinaesthetic response
Dreams, hallucinations and movement responses
How can movement be a determinant?
Seeing a movement is a case of practical knowledge
Seeing motor space
Ecological optics and the movements of others
Physiognomic perception
Inner and outer space
Moving towards the Other


Card 1
Card 2
Card 3
Card 4

Hermann Rorschach and the kinaesthetic response
The main contribution of Hermann Rorschach to the history of psychology and psychiatry was not the discovery that ambiguous pictures could be used for diagnostic purposes. The latter idea had occurred to several investigators before him.1   Instead, the real novelty of Rorschach’s approach lies in his insistence that people’s responses to the inkblots should be described not only in terms of what is being seen, but also in terms of how it is seen. Hence the most important things to note about any response are such things as: which part of the blot is being interpreted? is it a small or a big part? is the percept diffuse or structured? is it based on contours, or on the colours and shadings? is it a ”good” form or a ”bad” form?, and so on. This insight is nowadays represented in a four-fold way of classifying responses: manner of approach, determinant, content and originality. Rorschach does not himself use these terms, but it is convenient to reconstruct his ideas with the help of them.

The most original, and probably the most debated, of Hermann Rorschach’s ideas is that of movement as a determinant.  In his book,2   Rorschach defines the movement responses as

He stresses the point that not any seen movement is to be classified as a movement response. The essential question to ask is whether the movement is being felt, and not only seen. The ”feeling” which is needed is the kinaesthetic feeling which, according to Rorschach, belongs to a motor impulse to perform the movement which is seen. Hence, movement responses are often accompanied by a manifest tendency on part of the subject to imitate the seen movement.

From Rorschach’s definition also follows that ”a stone falling from a roof” is ordinarily not a movement interpretation. And even when a human being is seen as moving, the response need not always be determined by any kinaesthetic-feeling component. The alternative name for movement responses, ”kinaesthetic responses”, is therefore to be preferred. Importantly, ”a standing man” is as a rule a kinaesthetic response, although the man is not seen, or felt, as actually moving.
1 The best source concerning the pre-history of the Rorschach test is Ellenberger (1954, 1993).
2 Rorschach (1972) p. 25 (translation HM).

Dreams, hallucinations and movement responses
In his doctoral dissertation, which antedates his more famous book with nine years, Rorschach had described several cases of so-called ”kinaesthetic-optic” and ”optic-kinaesthetic reflex hallucinations”. These are cases where a visual hallucination is conditioned by kinaesthetic impressions or other bodily experiences, or a kinaesthetic/somatosensory hallucination is caused by visual stimuli. The first kind is exemplified when a schizophrenic patient who is performing rowing movements starts hallucinating first a lake, and then a whole fleet of war ships. When another schizophrenic subject reports that he feels that he has been transformed into the animal or inanimate object which he sees, we have an example of the second kind.3

Rorschach’s dissertation was partly inspired by the investigations of the Norwegian philosopher and psychologist John Mourly Vold, who had performed extensive and systematic studies of the influence of tactile and kinaesthetic stimuli on the contents of dreams. He found, for example, that people who have had their feet tied very often dream of walking, climbing and so on. These dreams usually involve the dreamer herself, but sometimes another person is instead seen climbing, and sometimes the dream just contains a ”climbable” object.4   Mourly Vold himself also extended his research and theorising to cases of hallucinations.5

Both Mourly Vold and Rorschach point out the similarity between these optic-kinaesthetic and kinaesthetic-optic hallucinations or dreams on the one hand, and everyday cases of connections between kinaesthetic, somatosensory and visual experiences on the other. Rorschach gives several such everyday examples:  when moving a limb in the dark, we  often vividly visualise it; many people visualise an object better if they ”draw it in the air”; we tend to feel the postures and movements of people which we see depicted on the scene and in paintings. He ascribes an important role to ”association” in the genesis of such inter-sensory connections, including those underlying many cases of reflex hallucinations. However, Rorschach believes that associative principles cannot explain all reflex hallucinations; indeed, he reserves the term ”reflex hallucinations in the narrow sense” for the non-associative cases.6  For these, an innate coupling seems to be required.
3 Rorschach (1912), p. 374, 364.
4 Mourly Vold (1900), p. 838 f; cf Mourly Vold (1910/12), quoted in Rorschach (1912), p. 371f.
5 Mourly Vold (1900).
6 Rorschach (1912), p. 397; cf also Mourly Vold (1900), p. 861.

How can movement be a determinant?
Rorschach’ definition of the movement, or kinaesthetic, response led to a still unresolved controversy within the Rorschach test community. One main problem has to do with the concept of a determinant. In connection with his discussion of form as a determining factor, Rorschach seems to say that a form response is one which is determined by the real form of the (interpreted part of the) blot.7  In the same vein, John Exner defines ”determinant” in terms of “the blot features that have contributed to the formation of the percept”.8  However, this way of thinking leads to a blatant contradiction if one wants to speak of movement as a determinant. Obviously, the actual blot does not move, nor does it entertain any intention to move. And if — as Rorschach says — kinaesthetic engrams (memory traces) are active when a subject sees a movement in a static blot, these engrams stem from the subject’s brain, not from the blot.

How is this difficult conceptual situation to be remedied? One possible way out is to move away, in theory and/or in praxis, from Rorschach’s statement that movement or kinaesthesia is a determining factor, and instead simply use ”Movement” as a kind of content category. However, it then becomes difficult to draw a non-arbitrary line between those seen movements which are ”Movement” interpretations and those which are not — not to speak about the corresponding problem for seen non-moving objects. At least, no clear reason remains to draw the lines where Rorschach once draw them.

A second option is to try to rethink the concept of a determinant. Perhaps we could relax the criterion that it must be a stimulus property? There are indeed strong evidence in Rorschach’s own text that when speaking about the determining factors behind the responses, he was actually thinking of the mental processes by means of which we try to make our percepts fit the stimulus properties. And Exner clearly steps away from stimulus properties when he uses the presence of emphatic identification as the basic criterion for the scoring of ”Movement”. However, Exner has dropped the specifically kinaesthetic component, because of the methodological problems involved in assessing the presence and importance of the ”kinaesthetic memory experiences” which Rorschach postulated. This means that in Exner’s system, ”Gloomy mood” can be classified as a movement response.
7 Rorschach (1921), p. 22.
8 Exner (1993), p. 103.

Seeing a movement is a case of practical knowledge
Hermann Rorschach’s main teacher was the famous Swiss psychiatrist Eugen Bleuler, and from Bleuler he also got his pre-Gestalt, associationist psychology. Had he lived, speculates his biographer,9  Rorschach would have developed his theory in line with phenomenology and Gestalt psychology. As it turned out, he left it to others to lay a modern foundation for his method. Until this has been done, the movement responses will remain a riddle.

The associationist doctrine that kinaesthetic experiences are necessarily involved in our knowledge of our own movements has since long been abandoned. In psychology the turning point was the research on ”imageless thought” by Ach and other psychologists of the Würzburg school around 1900, research which was partly inspired by phenomenology and which was followed up and deepened by the Gestalt psychologists. Within analytical philosophy, the theory of kinaesthetic impressions seems to have lived on until the later Wittgenstein gave it a final blow.10

Can we define ”kinaesthetic response” without postulating kinaesthetic impressions? Let us first remember that according to Rorschach, a subject who gives a movement response often shows a tendency to imitate the seen movement. He thinks that this should be regarded as a sign that the essential kinaesthetic experiences are present. An alternative, non-associationist description is that both the tendency to move and the kinaesthetic imagery (if either is present) are contingent expressions of the subject’s practical, perceptually based apprehension of which movement is performed. ”Practical apprehension” is here simply the intentional component of practical (procedural) knowledge. As Ryle says, we know how to swim without being able to describe it; as modern psychologists say, we have non-declarative, procedural knowledge of swimming. Subtract the conditions of truth and justification from this knowledge; what remains is the act of practical apprehension. And if such an act is based on perception, we have a perceptual, practical apprehension of a movement — a practical perception.

Thus, a kinaesthetic response in the Rorschach can be defined as a practical perception of a motor activity — a response in which the subject sees how to do what another person does. Some refinements are needed, however.
9 Ellenberger (1993), p. 230. Rorschach died less than a year after the publication of his book.
10 Wittgenstein (1953).

Seeing motor space
Not all practical perceptions of movement involve other people (or other animals). Indeed, the most common kind is that which occurs when we see how to do something, for example how to reach an object or how to walk to avoid an obstacle. In these cases, the immediate expression of our knowledge is not the imitation of an act, but the performance of the act itself.

When one sees what to do, or what is being done, which is the role of vision in the process? Not all such knowledge depends on the visual identification of the subject’s own movements, either concurrently or at any earlier time. The fact that very young infants know how to turn their eyes towards objects, and how to grasp for them, clearly illustrates the point. This point can also be expressed by saying that innate visuo-motor couplings play an essential role in perception-action chains. The well-known research of Meltzoff and Moore and others on imitation in very young infants is another example.11  The main conclusion to draw from this research is that the infant often has practical, perceptual knowledge about which act is being performed, although it does not see, and has never seen, itself perform the same act.12

There is an obvious connection here with the theory of the body schema as formulated by Merleau-Ponty.13  A major part of our knowledge of our own body has a practical character; for example, we know in the practical sense where to scratch the skin when it itches. Such knowledge is independent of vision not only because it works without visual input, but also in the sense that what is known it is not primarily a visual position. The body schema is  nothing but this fundamental system of practical knowledge. According to Merleau-Ponty the body schema — as a system of practical apprehensions — is essentially involved when one imitates another person. He also makes the point that imitation does not presuppose that the subject can see his own movements.14  It is therefore, in a certain sense, immediate.
11 Meltzoff & Moore (1995).
12 The methods and results of Meltzoff and Moore have been subject to criticism. See for example Heimann (in press). My main conclusion is however not threatened by this criticism. Cf. also Wyrwicka (1996).
13 Merleau-Ponty (1962). For my choice of the term ”body schema”  rather than ”body image” cf Gallagher (1995).
14 Merleau-Ponty (1962) pp. 352.

Ecological optics and the movements of others
During the last decade, some authors have noticed the connections between Merleau-Ponty’s phenomenology, research on imitation, and the ecological optics of James and Eleanor Gibson. I want to develop this theme here.

Gibson’s concept of affordance is naturally interpreted in terms of practical knowledge. In other words, what happens to an animal when it sees a plant as having the affordance being edible can be described by saying that the visual system makes the organism prepared to eat the plant. And Gibson’s point that perception of affordances is usually direct can be read as saying that there is no need for any cognitive representation to intervene between the ambient optic array and the practical apprehension of an affordance.15

There are a number of social affordances.16  An approaching enemy can, and should, be seen as somebody to flee from, while an approaching conspecific of the opposite sex may sometimes be seen as somebody to mate.  There are also affordances of being somebody to co-operate with  in specific ways.

One distinguishing mark of social interactions is that the required action depends on what is predicted about the partner(s). Such a prediction of a partner’s behaviour can, in principle, be made using two very different mechanisms. One may use the same method as when judging the movements of an inanimate object, which essentially reduces to extrapolations from observed movements (as for example in intercepting a thrown ball). Or, which is usually much more effective, one may try to see from the partner’s  perspective what the partner intends to do. A heavyweight fighter certainly stays upright longer if he tries to sense the other guy’s intentions than if he calculates time-to-contact with the approaching fist.

I suggest that the perception of social affordances involves the translation of visual input into two motor patterns, since the practical knowledge how to act depends on the practical knowledge how the other person acts. Only because the fighter knows practically what the other guy intends to do, can he avoid its consequences. If this description is correct, manifest imitation is just the special case where the two seen patterns are the same. The basic mechanism is much more general and has high ecological validity apart from imitation.
15 Cf Redström (1998).
16 J. Gibson (1979), p. 42; see also Bruce, Green & Georgeson (1996), ch. 16.

Physiognomic perception
Before we go deeper into the philosophical aspects of our theme, I want to point to another psychological and psychopathological field of relevance for it. I am referring to the existence of so-called physiognomic perception, of which at least a part may be described as misplaced practical apprehension of motor intentions. Hermann Rorschach points out that several test subjects seem to see — and ”feel” — human-like movements or motor attitudes in objects such as trees, household artefacts, etc. For example, a tree can be seen as standing proudly erect, or as stretching their branches towards another tree. Now, this way of seeing things is certainly not restricted to the test situation. People vary a lot with respect to how strong their tendency to this kind of ”perceptual animism” is.17  The same holds for the tendency to perceive faces and facial expressions in objects.18  The pioneer in developmental psychology, Heinz Werner, tells us that physiognomic perception is more common in young people,19  and the psychopathologist Klaus Conrad describes how the initial phases of confusional states are often characterised by a loosening of ordinary object perception and the intrusion of more and more physiognomic material.20

A natural explanation of these phenomena is the biological one. It is very important for the survival of organisms that they have a low threshold for the perception of other living creatures, so that they will be able to flee from aggressors in time. In this perspective, the sensitivity to movements of the peripheral part of the retina can be classified among the relevant facts. Also, a low threshold for physiognomic perception helps intraspecific co-operation to work smoothly. The price to be paid is a lot of misplaced perceptions.

However, I suspect that this explanation may be too simple, and that the ease with which physiognomic perception of inanimate objects can be initiated reflects that it may have a even more fundamental perceptual function than warning us about the possible presence of other creatures. I will next speculate a little about the nature of our perception of space.
17 Some (including the present author) can very easily put themselves in an attitude where almost everything is seen under such an aspect. For a marvellous description of how the world can seem to a person who is sensitive to physiognomic traits, cf Lingis (1996).
18 An interesting report on another case of high sensistivity in this respect can be found in the work of Franz From, who was a close collaborator of the famous Gestalt psychologist Edgar Rubin. Cf From (1971), pp. 66f.
19 Werner (1961).
20 Conrad (1960), pp. 385ff.

Inner and outer space
The idea that our apprehension of external space has a motor component is certainly not new.21  Merleau-Ponty has given one of the fullest formulations of it. Our perception of close external space, he says, is largely practical in nature: we see how to do in order to reach an objects, to avoid it, or to grasp it. Furthermore, our practical knowledge how to reach a certain object in our immediate visible environment is usually direct: we directly know how to move our arm to catch a close visible object. This is quite analogous to our direct knowledge where to scratch when it itches. Hence the body schema includes not only the body, but also a part of the environment. Merleau-Ponty’s highly original and suggestive comments on the nature of perceived depth must be read with these theses in mind.22

But what is then the status of distant space? Certainly, the horizon is not within one’s reach, so how can the perception of the horizon be analysed in terms of practical apprehension? The solution lies, I think, in Merleau-Ponty’s analysis of the role of the body schema in imitation. In the following excerpt he gives his most explicit definition of his concept of body schema:

I take this to mean that external space is given to us as a practical system of possible movements from different positions, where each position, in turn, is a point in the same space. Distant objects and the horizon are apprehended as reachable in steps. Seeing inanimate objects as having intentions is simply one useful way of imagining starting points for some of these steps through space. Hence the abundance of physiognomic perception.
21 When Berkeley says that space perception is founded on the sense of touch, one possible interpretation is that what is seen  in spatial perception is not a set of purely visual qualities but a set of motor possibilities. See especially Berkeley 1901 (1709), pp. 148ff.
22 Merleau-Ponty (1962), pp. 254ff.
23  Merleau-Ponty (1962), pp. 141-2. On the term ”body image”, cf above, note 13.

Moving towards the Other
Let us now return to the main theme of this paper. The reader may already have guessed that I want to define a kinaesthetic interpretation in the Rorschach test as a response which is based on the usual mechanism for practical perception of motor intentions, as described above. This definition also clarifies the wide sense in which all movement interpretations can be said to be ”form responses”.24  It is well known that perception of real human and animate movement is based on high level optical invariants, as is for example illustrated in Gunnar Johansson’s famous experiments.25  As Gibson points out, there are also event-specifying invariants in momentary or frozen displays such as ordinary (static) pictures.26  These invariants have not been studied to anything like the extent of the studies of real motion perception. But it is obvious that the optical invariants for specifying human motor intentions in a still picture are also very high level and not easily describable in terms of elementary shapes.27  Of course, Rorschach-like displays would offer great opportunities to study them better!

In the meantime we have to rest content with saying that kinaesthetic interpretations are  a kind of ”form” responses, although they are based on shape information of very specific kinds and use this information in a very special way. If an interpretation involving movement is instead based on the mechanisms which we ordinarily use in the identification of the shapes of close inanimate objects, it is a form response in the usual, narrow sense.

My final, philosophical suggestion is that the above analysis of the way we perceive motor intentions allows for the possibility that these perceptions are independent of our objectivating perceptions of close spatial structures. This gives a clear sense in which the perception of other subjects may be prior to the perception of objects: extracting the invariants necessary for the former achievement need not presuppose extracting the (lower-level) invariants needed for the latter. If combined with the Gibsonian view that practical perception is usually direct, the analysis also leads to the conclusion that our knowledge of other minds may be direct. The latter thesis can then be seen to be compatible with a thoroughly naturalistic epistemology.28
24  Bohm (1972), p. 46.
25 Cf. Bruce, Green & Georgeson, pp. 328ff.
26 Gibson (1979), p. 294.
27 This, by the way, explains the fact noted already by Rorschach that kinaesthetic responses in the Rorschach test often bridge otherwise clear contour ”gaps”.
28  Incidentally, this conclusion is very similar to the main point in Malmgren (1976). Cf also Malmgren (1983).

I want to thank Gösta Fröbärj, Susanna Lundqvist, Filip Radovic and Johan Redström for valuable comments and Linnea Malmgren for the blots!

Berkeley, G., An Essay towards a New Theory of Vision (1709, in: A. Campbell Fraser ed., The works of George Berkeley, Vol. I, Oxford 1901).
Bohm, E., Lehrbuch der Rorschach-Psychodiagnostik. 4 Aufl., Huber 1972.
Bruce, V., Green, P.R., Georgeson, M.A., Visual Perception. Psychology Press 1996.
Conrad, K., Die symptomatischen Psychosen. In: Gruhle et al, Psychiatrie der Gegenwart, Bd. II. Springer 1960, 369-436.
Ellenberger, H., The life and work of Hermann Rorschach (1884-1922). Bulletin of the Menninger Clinic, 18 (1954), 172-219. Reprinted many times, e.g. in Beyond the Unconscious. Essays by Henri F. Ellenberger. Ed. M. Micale, Princeton U.P. 1993.
Exner, J., The Rorschach: A Comprehensive System. Part I.  Wiley 1993.
From, F., Perception of Other People. Columbia U.P. 1971.
Gallagher, S., Body schema and intentionality. In: Bermúdez, J.L. (ed), The Body and the Self. MIT Press 1995, 225-44.
Gibson, E.J., Ontogenesis of the perceived self. In: U. Neisser (ed.), The Perceived Self. Cambridge U.P. 1993, 25-43.
Gibson, J.J., The Ecological Approach to Visual Perception. Erlbaum 1979, 1986.
Heimann, M., Imitation in neonates, in older infants, and in children with autism: feedback to theory. In: Bråten, S. (ed.), Intersubjective Communication and Emotion in Early Ontogeny. Cambridge U.P. (in press).
Kramer, R., The Rorschach M response: a return to its roots. J. of Personality Assessment, 57 (1991), 30-6.
Lindqvist, G., Malmgren, H., Classification and Diagnosis in Organic Psychiatry (= Acta Psychiatrica Scandinavica 88, Suppl. 373, 1993).
Lingis, A., The body postured and dissolute. In: Fóti, V.M. (ed.), Merleau-Ponty: Difference, Materiality, Painting. Humanities Press 1996, 60-71.
Malmgren, H., Immediate knowledge of other minds. Theoria 42, 1976, 189-205.
Malmgren, H.,  Immediate Knowledge: A study in G.E. Moore’s epistemology. Doxa, Lund 1983.
Meltzoff, A.N., Moore, M.K., Infants’ understanding of people and things: from body imitation to folk psychology. In: Bermúdez, J.L. (ed), The Body and the Self. MIT Press 1995, 43-70.
Merleau-Ponty, M., Phénomenologie de la Perception (Paris 1945). English translation Phenomenology of Perception (Routledge 1962).
Mourly Vold, J., Ueber ”Hallucinationen”, vorzüglich ”Gesichtshallucinationen”, auf der Grundlage von cutan-motorischen Zuständen und auf derjenigen von vergangenen Gesichts-Eindrücken. Allgemeine Zeitschr. für Psychiatrie, 57 (1900).
Mourly Vold, J., Über den Traum. Ed. O. Klemm. Barth, Leipzig 1910-12.
Redström, J., Affordances, information pickup, och direkt perception. Göteborg University, 1998 (mimeographed).
Rorschach, H., Psychodiagnostik. Eine Wahrnehmungsdiagnostische Experiment. Bircher, Bern 1921; later eds. Huber, Bern. All quotes from 9th ed (1972).
Rorschach, H., Über ”Reflexhallucinationen” und verwandte Erscheinungen. Zeitschr. für die gesamte Neurologie und Psychiatrie, 30 (1912), 357-400.
Werner, H., Comparative Psychology of Mental Development. New York 1961.
Wittgenstein, L., Philosophische Untersuchungen/Philosophical Investigations. Blackwell 1953.
Wyrwicka, W., Imitation in Human and Animal Behavior. Transaction Publ., 1996.

It would be of obvious interest to compare Hermann Rorschach’s ideas with a number of contemporary theories and empirical findings which I have not discussed in the poster. Pertinent examples are Kosslyn’s theory of mental imagery and the recent neuropsychological findings by Jeannerod and others about visuo-motor couplings in the brain.29,30  I intend to make these comparisons in a later, longer version of the paper.31
29 Kosslyn, S., Image and Brain. MIT Press 1994.
30 Jeannerod, M., The Cognitive Neuroscience of Action. Blackwell 1997.
31 (Note added in 2002) Malmgren, H., Rorschach's Idea of a "Movement" Response in the Light of Recent Philosophy and Psychology of Perception. Rorschachiana. Yearbook of the International Rorschach Society 24, 1-27. Hogrefe & Huber Publishers, 2000. Also available here.

Card 1

Two dancing gorillas can be seen in this Rorschach-like picture. (Thanks to Linnea Malmgren!)

Card 2

Is this two persons whose postures express discontent with the dance on the previous picture?

Card 3

This picture also invites to interpretation in terms of an emotional expression.

Card 4

1. One can see this as the roots of a tree.
2. Or as a the feet of a big bird, or a dinosaur.
If you now go back to the first interpretation, it may have aquired a ”physiognomic” character!

Top of page
Text page
Main page