15. Gesture. Acquisition of movement or meaning

The article presents the issue of the relation between acquisition of gesture movement and methods of defining meaning. The background of this consideration is placed in the common project of The Institute of Visual Research and Interaction (The Academy of Fine Arts) and The Institute of Theoretical and Applied Informatics of the Polish Academy of Sciences (Gliwice).
The project is focused on creating user interfaces which are based on gestures. The foundation of this work is the conception of combining informatic analysis (processing and classification of movement signals – ITAI Gliwice) with semantic analysis (meaning) of gestures (IVRI Katowice). The creation of methods of describing gesture meanings in communication between human and computer with particular consideration of two-sided interactions is essential. Those methods are based on the semantic field conception understood as a collection of meanings which are ascribed to a specific gesture
Considering the gesture topology created by David McNeill, a professor from The University of Chicago („Hand and mind. What Gestures Reveal about Thoughts”), who claims that gestures are an integral part of the language equally as words, phrases and sentences and belong to one system and his point of view that gestures like concepts, can show the process of human thinking – in conclusion we want to present a question – inverting the title – Can meaning tell us about movement?

Wiesław Gdowicz
Institute of Visual Research and Interaction Academy of Arts in Katowice Raciborska 37

Marta Więckowska
Institute of Visual Research and Interaction Academy of Arts in Katowice Raciborska 37

IV edition of the scientific seminar „User Interface – Kansei in practice”, organized by the Multimedia Department of the Polish-Japanese College of Technology in Warsaw, 30 May 2009 r

Copyright is held by the author/owner(s).
Kansei 2009, May 30, 2009, Warszawa, POLAND
ACM 1-xxxxxxxxxxxxxxxxxx.

Gesture, interface, semantic fields, semantic analysis.
ACM Classification Keywords
H.5.2 [User Interfaces]: Theory and Methods.
H.1.2 [User/Machine Systems]: Human information processing


The article presents the issue of the relation between acquisition of gesture movement and methods of defining meaning. The background of this consideration is placed in the common project of The Institute of Visual Research and Interaction (The Academy of Fine Arts) and The Institute of Theoretical and Applied Informatics of the Polish Academy of Sciences (Gliwice).
The project is focused on creating user interfaces which are based on gestures. The foundation of this work is the conception of combining informatic analysis (processing and classification of movement signals – ITAI Gliwice) with semantic analysis (meaning) of gestures (IVRI Katowice). The creation of methods of describing gesture meanings in communication between human and computer with particular consideration of two-sided interactions is essential. Those methods are based on the semantic field conception understood as a collection of meanings which are ascribed to a specific gesture
Considering the gesture topology created by David McNeill, a professor from The University of Chicago („Hand and mind. What Gestures Reveal about Thoughts”), who claims that gestures are an integral part of the language equally as words, phrases and sentences and belong to one system and his point of view that gestures like concepts, can show the process of human thinking – in conclusion we want to present a question – inverting the title – Can meaning tell us about movement?


Gestures are a significant part of communication between living beings, especially between human beings. They are an element of interpersonal communications which merge movement activity space with meaning space. That is why this problem may be considered from two points of view:
(a) technical and informatic methods of recording movement in space
(b) methods of reading and interpreting gestures .
Communication between computer and a human being should take into consideration the understanding of technical aspects of a device and the interpersonal nature of interpersonal communication of human beings.
The gesture which is a movement that includes meaning and is a mediator or a sign between the communicating sides. A gesture is not only a mediator but also through it a definite idea gesture expressed. Gesture should be considered as a mediating and expressing element. These two gesture’s aspects – mediating and expressing – are inseparable gesture features. When a gesture is perceived the question arises: how it is mediating and expressing. From the intermediary point of view the gesture has a physical character connected with the perception of movement of an object in space. From the expressing point of view the gestures mental character connected with the process of signing ideas .
Preliminary analysis of gestures as an element of communications begins from the collection of information about the role gesture in existing communication systems. It will show how the gesture was used till now, how it was read and interoperated. The preliminary analysis is based on the following issues:
(a) Artificial mechanical languages and systems of information codes – most of gestures in this systems has a function of a system connected with others by a syntactic rule . The sign collection and syntactic rules between them can be here analyzed. Most of these languages and systems are limited by a concrete situation and are closed collections.
(b) The language and recording systems for communication with handicapped people – the American sign language and recording system CH. Blissa and gesture system which supports SIS – this issue is important above all because it includes very specific communication features, which do not appear very often.
(c) Kinetography – The System of Analyzing and Recording Movement Laban is a system and language for understanding, observing, describing and notating all forms of movement. The system consists in transforming the three-dimensional movement into a two-dimensional notation.
(d) Analysis of a film material with respect to gestures – (silent film) , the gesture transferring ideas by a film image is purposely prepared and consists of many elements. The multicomponent film gesture is a reflection of the structure of the content transferred. Linear and sequential film features show new aspects of gesture.
(e) Semiology – especially considering U. Eco’s work – a theoretical elaboration of the gesture meaning space which on the one hand puts it together with the language structure and on the other hand with widely considered cultural aspects.
The collection of results of the preliminary analysis enables to describe the gesture character/features in relation to it’s physical use and it’s meaning functions as well.

1 Example of gesture analysis in Charlie’s film „Burlesque on Carmen”.

Gesture and sign

Not every movement is a gesture. This means that in the boundless world of human movements exists a specific movement collection, which can be called gesture. What kind of collection is it? What kind of conditions has been included in a movement to be a part of such a collection? What are the criteria of such a affinity? Are they constant or changeable? Does the belonging to the collection of gestures to is unconditional, i. e. if a movement becomes a gesture, is it a gesture forever?

These few questions are considered to be essential in describing gestures.
1. Gesture being a movement which has a meaning is a mediator, i.e. a sign between the communicating sides. Here the gesture is understand widely including all the context of intercession action. It can be said that in a gesture the hole space “between” the sides of communication is included. Another problem is the distance between the sides of communication : the same gesture changes its meaning with the variation of the distance between the sender and receiver.
2. A gesture does not only mediate, but it also expresses a specific idea. Gesture as an intermediary factor is a sign and thereby all the features of a sign can be also used here as well.
3. Gesture should be considered to be an intermediary and expressing factor . This statement emphasizes the link between these two gesture dimensions and means that the gesture in the intermediary dimension can not be considered without the expressing dimension and vice versa.
4. From intermediary point of view, a gesture is a physical object connected with the perceiving of an object moving in space. All what is connected with the gesture perception and the ways of perceiving it.
5. From the expressing point of view, a gesture is mental object connected with the signing of ideas. The analysis of the hole information process, the personal , cultural and social conditions. Conventions and gesture systems.

What do we know?

The nonverbal communication research began in the twenties century when Ruesch and Kees (18) have analyzed the semantic value. This research brought them to the statement: „the understanding of an announcement is not only a verbatim decoding of an information but also the grasping of its lateral values”. In their theory, the act of communication is understand multimodally. This means that information functions are also carried out by the position of the body, the glance or the space distance. The act of communication researches concern three types of nonverbal communication: the paralanguage, kinesics and proxemics. The gesture classification was taken from the work of David Mc Neill at the University of Chichago (19). It contained five basic type of gesture: (1) deictic gestures (2) beats (3) kohesive gestures(4) iconic gestures and (5) metaphoric gestures.

Most of the discussions and reflections concerning the communication understand as an act of interpersonal communication emphasize first of all its verbal character. It is probably the consequence of taking the speech act theory of J.L. Austin toorigidly (20). Nonverbal elements of speech acts were considered as non-relevant in the language space of the speech background. The development of devices which react to the movement inside the registered image bring noticeable gestures as important elements of visual communication. The focusing on nonverbal elements of communication involved specific consequences. One of them is the attempt to describe and classify gestures as being elements of the language structure. The latest interdisciplinary research (11) conducted inter alia by the team of the Experimental Psychology Department of KUL led by prof. Piotr Francuz shows how important role play visual codes in communication . Attempts have been made to seek an universal visual code, which could be the base of recognizing the perceived object as specific figure. They brought very good results. It confirms that an essential role in communication plays the perceiving category. It enables the transfer of the image content by perceiving shape and figured things. The researches claim that the shape and its specific properties organize the process of identification, interpretation and transformation of sensory data and make it without basing on lingual structures.

An attempt of creating a gesture classification was described in „Visual and Linguistic Information in Gesture Classification „(2). The review of psychology and linguistics gesture classification allows to accept a conception which divides gestures into four groups: (a) deictic gestures – gestures placed in a physical space, refer to units and a localization or to space which has been earlier marked as a relation of idea or conception; (b) iconic gestures – gestures that show selected action features or describe an occurrence, for example: ‘we were driving all over the city’ – showing the route by a hand movement; (c) metaphoric gestures – gestures that are also representative but do not show physical elements, the gesture shape comes from common metaphors, for example ‘it is always repeated’ – round shape hands movements. (d) beat gestures – those gestures are connected with movements like baton movements. The shape of the gesture does not change according to the speech content. Beats play a pragmatic role appearing as a comment or support…Speakers emphasize important elements of their speech by downward hand movements using regular gestures.
Basing on McNeill’s work, the authors suggest not thinking of gestures as separated categories but as a combination of various features. Researches have proved that there is no strictly specified gesture collection in which movement trajectory would define every gesture group. Gestures are defined in category rather by the role they play in the conversation than by specific hand movements. Relations from two experiments are presented which had to give the answer for such questions as: (A) In what range our gesture perception depends on visual analysis or linguistic features? (B) Is it possible to built a system of gesture classification using only linguistic elements? What kind of linguistic features deliver more information to this aim?

Ad different approach presents the paper „Towards Natural Gesture Synthesis …”(4). This research has been aimed mainly at the creation of more natural gestures. The researchers have analyzed video materials in that tv presenters use specific gestures. The material was so prepared that an automatic analysis by decoding device is possible. The decoder breaks the gesture and language timelines into elements connected with speech vocalization (preparation, stroke, hold, i.e. means g-phase). The gesture profile is created by adding supplementary annotations collected from other sources. The next step is the elimination of unimportant words and sounds from the content. Material prepared in this way is the foundation for creating an animation script. A distinctive feature of this approach is a strict union of a gesture as a movement with speech dynamics and the internal lexical structure.
Researches on the methods of gesture data analisys (1) are conducted in such a way that the result can be used to multimedial applications that provide interaction with the virtual environment. The understanding of uninterrupted gestures is a part of broader research which allows to make the analysis and adaptation of natural gestures data for computer systems. The primary aim is the research of most natural interactions where natural means that human possibilities and limitations are not adopted to the computer system, but the latter is adopted to those possibilities and limitations. In this case, the research consists in recording movements by movement sensors placed in the overall. The research results should give answers to questions concerning the kind of performing gestures during conveying a description and the problem of characterizing the gesture in such way that computer understand it.

Almost all the works considering the semantic properties of gesture combine the expression of gesture movements with speech articulation and treat this elements as inseparable. This makes the understanding of the role of meaning in the entire gesture conveying difficult. The understanding of the correlation between gesture meaning as a sign of conveying ideas with the movement description in space is also difficult.


The project which has been started by Institute of Visual Research and Interaction (Academy of Art in Katowice) is based on the creation of methods of describing gesture meanings in communication between human and computer with a particular consideration of two-sided interactions.
The methods of describing gesture meanings are based on the semantic field conception. The semantic field is understand as a collection of meanings which are ascribed to a specific gesture. One gesture is described by many meanings and in the semantic field they are organized hierarchically. The larger is the appearance probability of a specific meaning, the higher is the meaning in the hierarchy of all ascribed meanings. The highest meaning in the hierarchy can be found as the most natural meaning of the gesture.

2 Semantic field

The problem area

The analysis of gesture interpretations in various disciplines. The results of a preliminary analysis allow to define the gesture character in relation to its physical use and to its meaning function. They allow also to make a theoretical study of gesture as a sign of idea from the epistemology side.
2. Method of gesture meaning description. Conception of semantic field is a new way of treating gesture meanings. It is based on data collected during the semantic tests – four pre-gesture collections which are analyzing by statistic procedure which is necessary to specify the probability degree of the meaning occurrence of the same pre-gesture in every collection. In the next step, the places in the probability hierarchy of the meaning of a gesture in every collection are compared.
3. Gesture recording. Not all performed body movements are gestures. The collection of possible body positions and movements are named pre-gestures and are arranged in two morphological tables. The morphological table is a collection of views of body elements collated with a collection of movements those elements; the morphological table 2 is a collection of views of one one body element (for example the right hand) collated with others body elements (for example the left hand). This preliminary recording is a base for selection of pre-gestures which will be subject to semantic tests. Gesture recording is the essential element of the project and includes preliminary recording (creating morphological tables) and proper recording (creating pre-gesture collections). The proper recording comprise prepared material for semantic tests, i.e. the four pre-gesture collections.

3 Gesture morphology – gesture card

4. Semantic tests. Semantic tests enable to evaluate the gesture semantic field of the four collections of pre-gestures: (a) collection of pre-gestures moving in real space; (b) collection of static pre-gestures; (c) collection of moving pre-gestures; (d) collection of pre-gestures moving in a virtual space . The tests are aimed at the collection of all possible meanings of pre-gestures . Tests are conducted on a representative group of 30 people selected on a stratified sampling for every pre-gesture collection. The participants were asked to describe the presented pre-gesture by three words. The final results will show the most probable meaning of every pre-gesture from the four collections and the probability distribution of all the other meanings. The results are analyzed statistically including correlation methods and a path-analytic approach. A special application has been prepared (active pdf document) that makes it possible to conduct semantic tests for the b and c collections by internet. Here the statistical sample is accidental and depends on the visitors and e-mail responders.

4 Example of table of semantic tests

5. Gesture catalogues. Every catalogue will include a kart of gesture description, a gesture chart, a list of meanings, a description of the semantic field, a meaning hierarchy settlement insight every semantic field and a map of the semantic field.
6. Interface. Computer interpretation of human gestures. The main aim of the project is to create methods for defining gesture meanings. The interdisciplinary character of the project will appear in the application of the methods created to the conception of an interface controlled by gestures. The interface registers selected hand gestures and transforms them to commands which control the application. This application will be used to conduct visual tests. An important element of this interface is the possibility of reading not only the users decision but also the physical features of this decision like reaction speed, rapidity and so on. These elements combined with naturalness of using hands to control the test process enable to analyze the user reaction very precisely.

Growth point

Supplementary analyses showed many different aspects connected with gestures which are outside the clear human – computer communication issue. One example are the conclusions of David McNeill who is a linguist and works at the University of Chicago. His researches were aimed at defining the functions and positions of gestures in the communication process and their relation to the language. The foundation if his theory is the statement that gestures are an integral part of the language and not only a background for it.
During the tests he performed the participants described to another person an earlier presented film. The participants were informed about the task in advance. Every story was record and analyzed the appearing of gestures and their relation to the words used. In one experiment the participants have described one cartoon episode – „Tweety and Sylvester”. Mc Neill have analyzed fragment of the description:
“and Tweety Bird runs and gets a bowling ba[ll and drops it down the drainpipe].”
This fragment of the description was accompanied by the gesture: two hands thrust downward at the chest level. It looks as if the hands were thrusting a big round object. The downward thrust was targeted precisely at the “it down” fragment.
This moment of the description is an example of the gesture and word integration: words “it down” plus the image of a downward thrust.

5 Two modes of communication

Gestures and language are two very different modes to pass information and this combination expresses the same idea unit using opposite semiotic measures: picture and sound. A gesture provides the imagery and is entirely a meaning , while in the language the global meaning consists of every word meanings which are based on the word morphology. The gesture shape is formed during speaking and is determined by specified meaning and by the context in which the gesture exists.
Speech and gesture are contrasting modes of communication and this is the base of the theory of gesture and language dialectic. The combination of gesture and language – two contradictions – is an unstable construction, which seeks out a situation in which both elements express the same idea . The main elements of dialectic are growth points, which are moments when speech and gesture are synchronous, when two opposite modes of idea are expressing the same idea unit. A growth Point appears during the fragment “it down,” plus the image of a downward thrust.

6 Growth Point

What is happening before and after this specific point? This process has a certain periodicity, i.e. the gesture live-time:
Preparation: the hands start to move up from the speaker’s lap to prepare for the downward thrust.
Storage: Then the hands stopped at the very end of saying “drops,”
Stroke: Next was the gesture stroke, the downward thrust itself ,timed exactly with “it down”
Storage of the actual gesture: Proper movement ended in the middle of “down,” hands freezing again in midair until the word is finished (the second underlining).
Restraction: Finally, the hands returned to rest (end of second underlining up to the right bracket).
Unpackaging – „unpackaging the meaning” – utterance constructing – finding the best form to express thoughts, to give the strictly defined meaning.
The two pauses reveal that the downward thrust was targeted precisely at the “it down” fragment: the downward thrust was withheld until the speech fragment could begin and was maintained, despite the lack of movement, until the fragment was completed.
The growth point cannot be understood without determining its relation to context. Context is created by words and gestures which appear before the growth point. Cartoon plot is based on Sylwester’s continuous attempts to catch Tweety. These two characters are opposite forces. The verb “drops” describes the downward bowl movement, but the downward thrust of the analyzing gesture omits this word. That is because this verb doesn’t describe the bowl’s role. The bowl expresses the antagonistic force against Sylwester, against his actions. The verb “drops” describes what Tweety has done, instead of describing the intentional meaning of the image. The meaning the speaker wanted to convey was not connected with dropping something, but with showing a bowl as an image of antagonism between the cartoon characters. In this regard, details of combing gesture and speech form a construction created by the speaker.

7 Unpackaging – „unpackaging the meaning”

During the discourse, the speaker seeks for the best form to express his thoughts. To achieve this, the speaker uses two completely different modes: image and word. By choosing the words “drops it down,” the speaker supplements the meaning of the utterance by adding a gesture which depicts the antagonistic force that was omitted in the linguistic mode. This moment is the end of the dialectic process between the gesture and the language. It stops exactly at the growth point, when the language and the gesture complement each other, and where the dynamism (gesture) meets stability (word).
There’s a reason for the periodicity of the gesture. The elements preceding the growth point form the utterance framework which is crowned by the growth point. Gestures which are dynamic and instantaneous “can” predict the linguistic content, and at the appropriate moment unite with words and fill up the meaning. The additional meaning was omitted in words, because it could be better expressed by gestures.
Gesture and speech are not only a message or communication mode, but begin the thinking process for speaking – gestures carry ideas to existence in the material world. This shows the embodiment of meaning into concrete enactments or material experiences. This conception says that the gesture, the actual motion of the gesture itself, is a dimension of meaning. Gesture in the growth point carries living meaning, which inhabits the word.
This elaborations are connected with the „material carrier”. The material carrier – an idea used by Vygotsky (1986) is a concept which says that the gesture is a dimension of thinking. It is a beginning of the abstract model, which points out the enormous imaginary power of materialization. The gesture is ”focused” on transferring the meaning into concrete existence and becomes the part of the speaker’s existence in this exact moment. The growth point is positioned as the initial thinking unit for speaking, from which a dynamic process of organization emerges. The Growth Point is connected with creating elements of utterance in real-time. The name of the growth point refers to the concept, where a specific moment of thinking for the speaking process exists.

Gestus est motus et figuratio

The most complex definition of gesture comes from the XII century. Hugon of saint Victor, the abbot of St. Victor’s, is the author of this definition. He left a rich legacy of literature achievements. “De institutione novitiorum” is a handbook which shows novices a way to eternal blessedness. Reaching it requires a lot of conditions to be fulfilled. One of them is living with discipline. Chapters from X to XXI describe this subject. In chapter X, we find the definition of discipline: “Discipline involves ordered movements of all limbs, correct disposition for every pose and in every action”. The author states that discipline is useful and that this usefulness comes from dialectic of what is inside and outside. External body movements express the movements of the soul, and vice versa. Opposite actions of what is visible and what is invisible are settles using discipline.

Chapter XII starts with the definition of gesture:

“Gestus est motus et figuratio membrorum corporis, ad omnem agendi et habendi modum”
“Gesture is the movement and the configuration (the form) of limbs, adapted for every action and every pose.”

The word „figuratio” can be translated as “configuration”, “form” or “mode”, and has possesses the meaning of “shaping”. It means that the gesture configures a group of limbs and reveals what is hidden in figuratio – form, meaning the movements of the soul. Figuratio also says that every gesture is perceived by other humans and that this fact gives sense to gestures. A gesture is a hint and a sign. Victor recalls a few more ideas, which are connected with the definition of gesture. He says about gesture propriety (or impropriety), as a sign in relation to the meaning; about the purposefulness of idea – the gesture aims to cause a certain pose and measure of idea – a basic moral issue – the moderation of gesture can be understood as a restriction. Finally, the idea of gesture modality is an element which changes and specifies the movement. Precise classification of gestures and their mutual relations and interactions is described in the other part of the handbook.

The Latin term „Flatus vocis” – introduced by medieval nominalists – means word, or “word expression”. “Flatus” alone means “wind blowing, breathe; metaphorically: inflation, pride”. “Vocis” – is “a voice, tone, sound”. In English, “expression” is defined as “make yourself understood”. Expressing means being understood – of course, by others In this context, “word expression” means making words understood. Signing expression means making signs understandable. This also concerns signs expressed as imagery. The breath, which is included in flatus, comes up from inside, more precisely from place where human meets the thing being experienced. Only this way is it possible to make oneself, a word or an image or gesture understood.
Flatus vocis – word expression – word.
Flatus – wind blowing, bellows used to kindle fire, breath, breathing, inflation, pride (metaphoric)
Vocis – voice, tone, sound
Flatus vocis – voice carrying – word expression
Word expression – express – making oneself understood Flatus vocis – making oneself an understood voice
Flatus pictus – imagery expression – pictus – image, drawing, ornament

Flatus pictus – making oneself an understood image.
Flatus gestus – gesture expression – gestus – gesture
Flatus gestus – making oneself an understood gesture


1. Wexelblat A. An Approach to Natural Gesture in Virtual Environments. MIT Media Laboratory Sept. 1995
2. Eisenstein J, Davis R. Visual and Linguistic Information in Gesture Classification. Computer Science and Artificial Intelligence Laboratory, MIT Cambridge USA
3. Rittschera J., Blakeb A., Robertsc S.J.. Towards the automatic analysis of complex human body motions. Image and Vision Computing 20 (2002) 905–916
4. Kipp M., Neff M., Kipp K.H., and Albrecht I.. Towards Natural Gesture Synthesis: Evaluating gesture units in a data-driven approach to gesture synthesis. Saarland University, Experimental Neuropsychology Unit, Germany.
5. Jarmołowicz E. Niewerbalne elementy aktów mowy. Institute of Linguistics, Adam Mickiewicz University Poznań, Investigationes Linguisticae, vol. XII, Poznan, December 2005
6. Ampel-Rudolf M.. Konwecjonalne gesty I mimika we współczesnej literaturze. Bulletin de la Société Polonaise de Linguistique, fasc. lv, 1999
7. Kessler, Hodges, Walker. Evaluation of the CyberGlove™ as a Whole Hand Input Device.
8. Eisenstein J; Davis R.. Natural Gesture in Descriptive Monologues. MIT Cambridge 2003
9. Eslambolchilar P., R. Murray-Smith. Model-Based, Multimodal Interaction In Document Browsing.
10. Aggarwal J.K.. Problems, ongoing research and future directions in motion research. Machine Vision and Applications (2003) 14: 199–201
11. Francuz P.. Obrazy w umyśle. Wydawnictwo Naukowe Scholar Warszawa 2007
12. Eco Umberto. Nieobecna struktura. Wydawnictwo KR Warszawa 1996
13. Shaughnessy J., Zechmeister E., Zechmeister J.. Metody badawcze w psychologii. Gdańskie Wydawnictwo Psychologiczne Gdańsk 2002
14. Polański Z.. Współczesne metody badań doświadczalnych. Wiedza Powszechna Warszawa 1978
15. Hall E. T.. Ukryty wymiar. PIW Warszawa 1976
16. Whorf B. L.. Język, myśl i rzeczywistość. PIW Warszawa 1982
17. Wittgestein L.. Tractatus logico-philosophicus. PWN Warszawa 2004
18. Ruesch J., Kess W.. Nonverbal Communication. Notes on the Perception of Human Relations. Berkley 1956
19. McNeill D.. Hand and Mind. The University of Chicago Press 1992.
20. Austin J. L.. Mówienie i poznawanie. PWN Warszawa 1993
21. McNeill D.., Gesture and Thought, University of Chicago, The Summer Institute on Verbal and Non-verbal Communication and the Biometrical Principle, Sept. 2-12, 2006, Vietri sul Mare (Italy), organized by Anna Esposito.
22. McNeill. D., Imagery for speaking, University of Chicago
23. Acta Linguistica Hafniensia green – 6/28/02 Gesture And Language Dialectic, David McNeill, University of Chicago
24. Vygotsky, L. S. (1987). Thought and language. (Edited and translated by E.Hanfmann & G. Vakar; revised and edited by A. Kozulin). Za Cambridge:MIT Press. za David McNeill, Imagery for speaking, University of Chicago
25. Schmitt J-C, Gest w średniowiecznej Europie. Oficyna Naukowa 2008