m o Rp h o g e Ne Sis o f m e a Ni n g 01

8/3/2019 m o Rp h o g e Ne Sis o f m e a Ni n g 01

1/83

M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 1 of 99

This is the html version of the file http://www.crea.polytechnique.fr/JeanPetitot/ArticlesPS/Petitot_MM.pdf .G o o g l e automatically generates html versions of documents as we crawl the web.Google is neither affiliated with the authors of this page nor responsible for its content.

These search terms have been highlighted: morphogenesis meaning petitotPage 1

M O RP H O G E NE SISO FM E A NI N Gby

Jean P ETITOTTranslated by Franson M ANJALIM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 2 of 99

Peter Lang, forthcomingPage 23TABLE OF CONTENTSTRANSLATORS INTRODUCTION ........................................................... 3PREFACE TO THE ENGLISH EDITION ..................................................... 9INTRODUCTION................................................................................. 1 3CHAPTER I ....................................................................................... 1 7PROBLEMATIC ASPECTS AND KEY ISSUES OF STRUCTURALISM.............. 1 71 . UNDERSTANDING oeSTRUCTURE.................................................... 1 72. MAIN TRENDS IN STRUCTURALISM: A BRIEF REVIEW ................... 2 2M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 3 of 992.1. THE APORIA OF ORGANIZATION IN KANT SCRITIQUE OF THE FACULTY OFTELEOLOGICALJUDGEMENT.....................................................................................................................22

2.2. STRUCTURALISM IN BIOLOGY ................................................................................242.3. GESTALT THEORY AND PHENOMENOLOGY ................................................................262.4. THE STATES OF AFFAIRS (SACHVERHALTE ) ................................................................292.5. STRUCTURALISM IN PHONOLOGY (GENERALITIES ).....................................................322.6. ACTANTIAL STRUCTURES AND CASE -GRAMMARS (GENERALITIES ) .............................342.7. SEMIO -NARRATIVE STRUCTURES (GENERALITIES ) ....................................................373 . THE PROBLEM OF FORMALIZING STRUCTURES ............................... 4 13.1. THE INTRINSIC LIMITS OF THE FORMALIST PERSPECTIVE ............................................413.2. THE TOPOLOGICAL A PRIORI AS THE CENTRAL THEORETICAL PROBLEM OFSTRUCTURALISM ..............................................................................................................433.3. DELEUZE S PROPOSAL FOR A SCHEMATISM OF STRUCTURE .........................................463.3.1. The symbolic realm.......................................................................................463.3.2. The criterion of locality or of position..............................................................463.3.3. The differential and the singular dimensions......................................................473.3.4. Differentiating and Differentiation ..................................................................48

3.3.5. The serial function........................................................................................493.3.6. The empty place...........................................................................................493.3.7. From the subject to practice...........................................................................504 . THE NECESSITY FOR A MORPHOLOGICAL GEOMETRY...................... 5 15 . THE PRINCIPLES OF CATASTROPHE THEORY .................................. 5 2Page 45.1. PHENOMENOLOGY AND OBJECTIVITY ......................................................................535.2. FOUR GUIDING PRINCIPLES ....................................................................................545.2.1. Phenomenological abduction..........................................................................54M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PM


2/83

http://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 4 of 995.2.2. Ontological neutrality and phenomenological reduction......................................555.2.3. The independence from the substratum.............................................................555.2.4. Hylemorphism..............................................................................................565.3. CT AS A MATHEMATICAL PHENOMENOLOGY ............................................................575.4. CRITIQUE OF LOGICISM .........................................................................................595.4.1. Extensionality / Intensionality ........................................................................595.4.2. Cinematics / Dynamics..................................................................................605.5. CENTRIFUGAL DYNAMICS ......................................................................................615.6. PHENOMENA AS MORPHOLOGIES ............................................................................625.7. THE LOCALITY PRINCIPLE ......................................................................................645.8. MATHEMATICS AND REALITY .................................................................................65CHAPTER II ...................................................................................... 6 7CATEGORICAL PERCEPTION AND TOPOLOGICAL SYNTAX....................... 6 7A DOUBLE APPLICATION OF MORPHODYNAMICAL MODELS TO THE DOUBLEARTICULATION OF LANGUAGE............................................................ 6 71 . PHONOLOGICAL CATEGORIZATIONS AS CRITICAL PHENOMENA ........ 6 71.1. CATEGORICAL PERCEPTION ...................................................................................671.1.1. Definition....................................................................................................671.1.2. Function .....................................................................................................691.1.3. General abstract situation .............................................................................69

1.1.4. Examples.....................................................................................................691.1.5. Specificity...................................................................................................711.1.6. Innateness...................................................................................................721.2. CONFLICTING INTERPRETATIONS ............................................................................731.2.1. The sensorial hypothesis................................................................................731.2.2. The reductionist hypothesis of feature-detectors................................................731.2.3. Criticism of the feature-detector hypothesis......................................................751.2.4. The oefoundational aporia of phonetics ...........................................................761.2.5. The a priori of categorical perception and of paradigmatic categorization..............771.2.6. Elements of catastrophist phonology ................................................................79(a) Discontinuous internal features and continuous external features .................................79(b) The discretization condition..............................................................................81(c) The mathematics of phenomenological abduction ...................................................84M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 5 of 99

Page 51.2.7. Precatastrophic interpretations of categorical perception...................................84(a) Prototype models ...........................................................................................84(b) Precatastrophic models ....................................................................................851.3. PRINCIPLES OF CATASTROPHIST MODELING .............................................................881.3.1. The qualitative type......................................................................................891.3.2. The criterion of structural stability..................................................................901.3.3. Genericity and transversality.........................................................................901.3.4. Finite determination, universal unfoldings and transverse models.........................901.3.5. Degrees of instability and stratifications...........................................................911.3.6. Normal forms and the theorem of classification..................................................911.3.7. Methodological rules for modeling ..................................................................921.4. FROM THE SECOND TO THE FIRST ARTICULATION OF LANGUAGE ..................................922 . ACTANTIAL SCHEMATISM ............................................................ 9 22.1. TOWARDS A PURE EToeIC LINGUISTICS ....................................................................932.2. FIVE LIMITATIONS OF THE CHOMSKYAN PARADIGM ..................................................982.3. THE PRIMACY OF ACTANTIAL RELATIONS ...............................................................1012.4. ACTANTIAL SCHEMATISM AND THE LOCALIST HYPOTHESIS .......................................1033 . TESNIERE'S NOTION OF VERBAL VALENCE ................................... 1043.1. THE GRAPHICAL ANALOGY OF CONNECTION ...........................................................1043.2. THE STEMMAS ...................................................................................................1053.3. THE PRINCIPLES OF THELMENTS.......................................................................1063.4. VERBAL NODE AND VALENCE ...............................................................................1084 . THE GRAMMATICAL REDUCTION OF STRUCTURAL SYNTAX............ 1094.1. CRITICISM OF THE GENERATIVIST EVIDENCE ..........................................................109


3/83

4.1.1. Competence and performance ......................................................................1094.1.2. The inadequacy of rewriting rules .................................................................110M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 6 of 994.1.3. Remarks on the transformational theory .........................................................1114.2. CASE GRAMMARS ..............................................................................................1124.2.1. The basis for case theory .............................................................................112

4.2.2. The difficulties of case grammars ..................................................................1181. Predicative and equative sentences......................................................................... 1182. The one-to-one selection of roles by cases .............................................................. 1193. The ambiguity of case forms................................................................................ 1194.3. JOHN ANDERSON 'S LOCALISM ..............................................................................1204.3.1. Multicase interpretation of roles ..................................................................1204.3.2. An aspect of the localist hypothesis ...............................................................1214.3.3. Localist grammars ......................................................................................122Page 64.4. RELATIONAL GRAMMARS ....................................................................................1234.4.1. Basic hypothesis.........................................................................................1244.4.2. Transformations and Relational hierarchy......................................................1254.4.3. Grammatical relations and actantial schemas .................................................1294.5. THE SYMBOLIC MISUNDERSTANDING OF VALENCE ..................................................1305 . THE SCENIC CONCEPTION OF CASE ROLES AND THE

MORPHODYNAMICAL SCHEMATISM................................................... 1325.1. THE DISTRIBUTION OF ACTANTIAL SEMANTISM .......................................................1325.2. THE RELATIVIZATION OF CASE ROLES TO SCENES ....................................................1365.3. TOWARDS THE LOCALIST HYPOTHESIS ...................................................................1405.4. STRUCTURAL SYNTAX AND CATASTROPHES ...........................................................1416 . THE LOCALIST HYPOTHESIS ....................................................... 1456.1. THE HISTORY OF THE LH......................................................................................1466.2. THE LH AND THE SPATIAL CONCEPTION .................................................................1496.3. THE PARALOGISM OF LA CATGORIE DES CAS ...........................................................151M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 7 of 99CHAPTER III ................................................................................... 153SEMIO-NARRATIVE STRUCTURES ...................................................... 1531 . INTRODUCTION......................................................................... 153

2 . THE SOURCES OF GREIMASIAN THEORY ...................................... 1573 . PANORAMA OF GREIMASIAN THEORY.......................................... 1613.1. THE GENERATIVE PATHWAY .................................................................................1623.2. FUNDAMENTAL SEMANTICS .................................................................................1623.2.1. The notion of seme......................................................................................1633.2.2. Semic categories ........................................................................................1643.2.3. Nuclear semes, classemes, sememes, lexemes and isotopies ..............................1643.3. FUNDAMENTAL SYNTAX AND THE SEMIOTIC SQUARE ...............................................1683.3.1. Semiotic pregnances and the substance of content............................................1683.3.2. The semiotic square....................................................................................1693.3.3. Logical inconsistency of the semiotic square...................................................1723.3.4. The aporia of privative opposition ................................................................1743.3.5. The syntax of operations ..............................................................................1763.4. ANTHROPOMORPHIC SYNTAX AND THE ACTANTIAL THEORY .....................................1783.4.1. The conversion ..........................................................................................1783.4.2. Actants and the actantial model ....................................................................178Page 73.4.3. The objects of value and their circulation.......................................................1823.4.4. Narrative statements and programmes, tests, and polemical structure .................1833.4.5. Topological syntax and the syntax of operators........................................................1883.5. THE GAP BETWEEN FUNDAMENTAL SYNTAX AND ANTHROPOMORPHIC SYNTAX ............1903.6. MODALIZATION AND ACTANTIAL ROLES .................................................................1913.7. oeERVIDICTION AND MANIPULATION .....................................................................192M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 8 of 99


4/83

3.8. DISCURSIVIZATION AND FIGURATIVIZATION ...........................................................1943.8.1. The function of nuclear semes: discursive configurations and themes ..................1943.8.2. Thematic roles and actors............................................................................1953.8.3. Figurativization..........................................................................................1963.8.4. Spatio-temporal programing..........................................................................1963.9. THE NARRATIVE SCHEMA ....................................................................................1974 . PAUL RICOEURS OBSERVATIONS................................................ 199

5 . SCHEMATIZATION OF THE UNDEFINABLES ................................... 2075.1. GREIMASIAN EPISTEMOLOGY ...............................................................................2075.2. THE UNDEFINABLES AS UNIVERSALS .....................................................................2095.3. THE FOUNDATIONAL APORIA OF THE FORM OF MEANING ..........................................2115.4. THE NECESSITY OF SCHEMATIZATION ....................................................................212CONCLUSION.................................................................................. 215BIBLIOGRAPHY ............................................................................... 217M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-FnPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 9 of 99Page 891

MORPHOGENESIS OF MEANINGJean P ETITOT 1Translated by Franson M ANJALI 2M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 10 of 991 Ecole des Hautes Etudes en Sciences Sociales, 54 bd Raspail, 75006 Paris, and CREA, EcolePolytechnique, 1 rue Descartes, 75005 Paris, [email protected] Nehru University, [email protected] 101

TRANSLATOR SINTRODUCTIONThe present work was originally written as part of Jean Petitots Thsedtatdefended in 1982. It was published in 1985 by the Presses Universitaires deFrance,Paris, in their series Formes Smiotiques under the title Morphogense duSens. Pourun schmatisme de la structure.The second part ofMorphogense du Senswaspublished in 1992 by the CNRS ditions, Paris, under the title Physique duSens.

The importance ofPetitots original French publication can be emphasizedon twocounts. Firstly, it provides a deep philosophical elaboration of Ren

Thoms CatastropheTheory (CT) proposed in the mid-seventies. In his preface to Morphogense du

Sens,Thom acknowledges that the theory which had generated great hopes within thescientificcommunity at the time of its launching ended up being merely oea set ofrecipes formodeling, or a tool-kit for applied mathematics. He notes that thanks to JeanPetitotswork


5/83

oethe philosophical project underlying the whole enterprisehas beenspecified, clarified, amplified, and above all restored to its rightful placewithin the grand philosophical and methodological tradition of thesciences, particularly the social sciences.Secondly, and as for the works significance in the social sciences, ThomM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PM

http://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 11 of 99

approvingly refers to Petitots catastrophist reworking of Jakobsonsstructuralphonology via the notion of oecategorical perception, to his catastrophistmodeling of theoelocalist interpretation of the case category which functions as a sort offulcrum betweensyntax and semantics (the localist idea has a history beginning from theByzantines,Maxime Planude and Theodore of Gaza, to Charles Fillmore and John Anderson,via

Louis Hjelmslev), and most importantly to his catastrophist schematization ofGreimasiantheory of semio-narrative structures.Morphogense du Sens was a seminal work which exerted a deep influence onthe different semio-linguistic schools: Greimass and Coquets Frenchschools, EcosItalian school in Bologna, Urbino, and San Marino, Brandts Danish school inAarhus,Wildgens German school in Bremen, and also Canadese schools in Montreal(PierreOuellet and Pierre Boudon) and Qubec (Gilles Ritchot and GatanDesmarais). It hasbecome a key reference and we think it is therefore a good thing to provide itsEnglishtranslation.

The present English version is strictly targeted to a oescientificreadership. As JeanPetitot says in his Foreword, the oecontinental philosophical digressionshave beendeliberately eliminated almost fully. In the process, those aspects of the bookthat hadmade it appear epoch-making in the mid- and late eighties may be foundwanting in the

English version, but the focusing on its scientific oehard-core may bemore attractive andPage 124advantageous, especially to those who are familiar with the dynamical modelingperspectives that have emerged in large numbers in the cognitive sciences ingeneralduring the nineties, even if its impact is yet to be felt on the generally slow-moving


6/83

intellectual horizon of Linguistics (and Semiotics) in a clearly discernible form.Jean Petitot told me he would prefer to see this pioneering work as a sort ofoeretrospective contribution to the ongoing trends in dynamical modeling,or as a kind ofreminder of a strong antecedent which was relatively original for the Anglo-Americanacademic world during its period of euphoria with the Chomsky-Fodor type offormalistcognitivism as well as with other forms of logicism, and also as something that iscapableM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 12 of 99

of providing certain fresh insights into the relatively new dynamical paradigmwhich hasblossomed under the aegis of the oeconnectionist research enterprise incognitive science.

The central issue dealt with in this book is that ofstructure. More precisely, with

the question of assigning aphysical and dynamical basis to structure inlinguistics andsemiotics. The classical problem with structure has always been to conciliate itsformalessence with its phenomenal filling-in, its discrete oeform with itscontinuous oematter (touse Hjelmslevian terms). For instance, the categories of linguistic structuralism,beginning with the phoneme, etc., are not conceived classically as naturalcategories, butmerely as conceptual ones which are projected onto the real world. Themethodologicalstrategy employed in this regard is to suggest that the structural unit,irrespective of whereit occurs, and particularly the phonemic unit, is a type subsuming one or morenaturaltokens, e.g., the phones. In a phonemic analysis, the differences between thephones areidentified as distinctive or not. Once the phonetic / phonemic differences areidentified,and the distinctive (phonemic) units established, the latter are arranged inparadigms, andare seen as being available for combinatory (syntagmatic) deployment. But thepoint that

is missed in this classical formalist perspective is that there are no abstractcategories innature; categories are largely mental products resulting from aprocess ofdiscretelydividing up the natural entities. These natural entities do not exist as such asdiscreteentities, but form part of a continuous substratum. Therefore, a formalization ofthe


7/83

structural categories that exist merely as constructs is handicapped by the factthat it leavesbehind the continuous and the natural substratum from which structure and itscategoriesinevitably emerge.Hence the importance of using dynamical models which can explain howqualitative discontinuities can emerge from the organization of the continuum,in such away that it can be categorized and discretized. Catastrophist models yielded thefirstexamples of such algorithms generating discontinuities. Using them, Petitotinterpretedthe qualitative and the privative oppositions that form the basis of

Jakobsons(phonological) distinctive feature analysis in terms of the catastrophes ofconflict andbifurcation respectively.Page 13M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 13 of 995

The question of syntactic structure presents an even more interesting picture.Chomskyan axiomatics did go beyond the earlier oeitem and arrangementapproach in thisdomain, by introducing a principle of generativity, essentially based in two setsof rules,those of recursivity and transformation. The apparent autonomy ofChomskys generativedevice in fact masks the rootedness of the syntax of natural languages in the

structures ofaction and perception, in other words, the partial analogy that exists betweenthe structureof language and the structure of the experienced external world.An investigation of the core grammatical structure of natural language revealsnotso much an infinite generativity of sentence structures as Chomsky had onceclaimed, butrather an auto-limitation imposed by the patterns of action in the external worldand itsperceptual reception by the language-user. Thoms first important

intervention inlinguistic theory was to question the oepure and simple idolatory of theformalist notion ofgenerativity, and to insist on the need for explaining the auto-limitation of thegenerativecapacity itself.It is here that a grammatical analysis must turn to some of the non-formalist(and


8/83

rather realist and semanticist) perspectives on the case-structures. Morespecifically, theactantial perspective of Lucien Tesnire and the localist theory adopted byHjelmslev,Anderson and Fillmore become relevant. Tesnire had, in hisoestemmatic analysis of thesentential syntax, granted centrality to the verbal node and regarded the nounphrases asoeactants that are oedependent on that central node. Sentence-meaning was understood,not as the resultant of a combinatorics of word-meanings, but as somethingconfigurationally available in a gestalt-like manner. It was composed holisticallywith theverb conveying the action part of the sentence, and the oeactants playingthe role ofparticipants in the action. Tesnire was explicit about the theatrical imagery (infact he

refers to oea little drama) while speaking of sentence-structure and itsmeaning. It is ofinterest for us to note here that such a view of the sentence and its meaningwas preciselywhat was proposed by the early Indian grammarians in whose verb-centeredanalysis theterm karaka is an exact equivalent of the Tesnierian oeactant. Andmoreover, forBhartrhari, comprehension of sentence-meaning is equated with a gestalt-likeperception,or citra-jaana (pictorial knowledge).

Hjlemslev too, pursuing his project of a pure structuralism arrived at aperspectivenot too distant from the above one. In his celebrated book La catgorie descas, afterpresenting a historical survey of various views on the case-category , heconcludes thatcase cannot be a logical category, but only a structural one. He fully embracesthe localistM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 14 of 99

hypothesis of the cases coming down to us from the Byzantine scholars referredto above

via the 19th century Kantian linguist Wllner. In the final analysis, the case isforHelmslev, a category that signifies spatial relations between two objects. Hedefines theserelations along three oedimensions, namely, Direction (Distancing andNearing),Subjectivity-Objectivity and Coherence (with or without contact).Page 146


9/83

Thom has applied CT to define the genesis of the grammatical (case) structuresfrom the actantial dynamics (derived from Tesnire) on a spatial substratum.

The set ofseven elementary catastrophes functions as the founding principle for thededuction of thegrammatical cases. As a characteristic example, Thom gives the followingschemata forthe Accusative case (or, the oeactantial graph for capture):S2S1 S1I

where S 1 and S 2 stand for the paths, in time, of the actants, and I the point ofintersectionwhere the sudden disappearance ofS 2 takes place.

The above actantial graph is just one of a list of 18 oearchetypalmorphologies that

Thom has proposed, which are derived from the set of elementary catastrophes.

Thesearchetypal morphologies show more finely the correspondence between thetopologicalgraphs and the case structures. Thoms topologico-dynamical analysis ofsyntaxsemanticsthus involves a synthesis of the actantial syntax, the case grammar and the ideaofmorphogenesis coming from CT. The main philosophical import of Thom'stheory isthat it retains an essential continuity between the physical and thephenomenologicalmodes of existence, something that the logicist approaches do not wish to do orareM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 15 of 99

incapable of doing. In the present case, the appearance of phenomenologicaldifference ispreceded by a physical process of differentiation of an initially continuous stateto yielddiscrete entities.

Thus CT allows to deduce the qualitatively differentiated case-structures from atopologico-dynamic physical substratum. It provides a principle of identifyingandcategorizing the finite set of core grammatical (case) structures which in the

natural worldappear as infinitely varied occurrences of physical or physically-based actions.The mainmerit ofPetitots work in this regard lies not in proposing the originalintuition of theconnection between the CT and case theory, but in meticulously establishing theplace and


10/83

the relevance of CT as a viable dynamical approach (what Petitot will renameas amorphodynamical approach), in contrast to the various formalist approaches,withincontemporary linguistic theory. Petitots subsequent researches haveestablished contactswith the dynamical approaches in linguistics present explicitly or implicitly in theworksof Per Aage Brandt, Leonard Talmy, Ronald Langacker, and George Lakoff. Butthenlinguistics is still to wake up to the fact that oemathematical linguisticsbased on a logicoalgebraicformalization which was fashionable during the fifties and sixties has virtuallyPage 157given in to the oemorphodynamical approach of the nineties that employsa sophisticated

mathematical topology that can better handle the inherently dynamical andstructuralcharacter of the core grammar of natural language.Petitots more recent work has focused on dynamic modeling in visualperception.An excellent paper which presents Petitots perspectives on grammar andvisualperception is oeMorphodynamics and Attractor Syntax: Constituency in VisualPerceptionand Cognitive Grammar that has appeared in Mind as Motion - Explorationsin theDynamics of Cognition edited by Robert F. Port and Timothy van Gelder (MIT

Press,1995). It gives a comprehensive picture ofPetitots morphodynamicalapproach which isnow very much part of the dynamical tradition of doing connectionist cognitivescience,on either side of the Atlantic. Petitot would like to see it as a sort of synthesisbetween onM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 16 of 99

the one hand the European theoretical traditions of gestalt theory andstructuralism, and on

the other the American traditions of Cognitive Linguistics and the dynamicalmathematicalmodeling towards a connectionist AI.With regard to the analysis of semio-narrative structures (which comes downfromV. Propp to A.-J. Greimas via C. Lvi-Strauss), Jean Petitot's attempt has beentotheoretically develop the inherent topological potential of the semiotic square byapplying


11/83

CT. This he does by providing a schematization of Greimass structures ofelementarysignification and a catastrophist interpretation of the latters actantial modelof narrativestructure. Applying the theory on Greimas's model, Petitot suggests that therelationsassociated with the qualitative and privative oppositions of the semiotic squarecould beschematized by means of the catastrophe of Conflict of minimal complexity andthat ofBifurcation of minimal complexity respectively. This shift, he thinks is in tunewith thetopological potential of the square, and involves the abandonment of a logico-combinatorymethod which is not suitable for a method which must explain the emergence ofthestructure from a physical substratum. The main merit claimed for the

catastrophist modelin narrative semiotics is that it can schematize the oeundefinableconcepts of the previousformalist framework. The oemorphogenesis of the square can bemodeled as aoeprocession of elementary catastrophes. At a more complex level, theentire oecanonicalformula of narrative structures as proposed by Lvi-Strauss can beunderstood in termsof the schemas for two coupled qualitative oppositions, represented by aoedouble cusp

(which is an intricated singularity).The "conversion" that gives rise to the Greimasian actantial model from thesyntactic operations on the content values is seen in terms of the actantialgraphsassociated with the elementary catastrophes. For example, Petitot shows thattheconversion S O S O (i.e., a state of disjunction between the Subjectand theObject-of-value becoming a state of conjunction between the Subject and theObject-ofvalue)can be described by means of the actantial graph of oecapture. Asregards theintentional and/or metapsychological dimension which defines the Subject-ObjectPage 168M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 17 of 99

relationship of the interaction, Petitot reminds us that Thoms archetypalmorphologies are


12/83

indeed actantial schemas deeply rooted in the behavioral structures of livingbeings.Let me conclude with a personal note. This translation has taken a long period ofgestation. It began as something of a hobby during a stay at Maison delAllemagne, CitUniversitaire, Paris, while pursuing post-doctoral studies in Linguistics at theSorbonne.Subsequently, it became a very serious endeavour, with constantencouragement from

Jean Petitot. But eventually, we let it grow from being a mere translation into arevised and even a oerevisited version.I must acknowledge sources of material support this translation project hasreceived at various points during the last ten years: Maison des Sciences delHomme,Paris, Indian Council of Social Science Research, New Delhi, and CNRS, Paris. Iwouldlike to particularly thank Monsieur Maurice Aymard, Administrator of the Maison

desSciences de lHomme for the faith he posed in me. The most concertedcollaborativeeffort went into the making of this version of the book during my stay in 1997 atMaisonSuger situated in the throbbing heart of Paris. It finally looked like nearingcompletionduring Jean Petitots visit to the Indian Institute of Advanced Study, Shimla.It has been agreat pleasure working with him.Shimla, July 1999M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PM

http://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 18 of 99Page 17

PREFACE TO THE ENGLISH EDITIONThis English version of oeMorphogense du Sens looks more like aoerevisitededition than just a simple translation. Indeed, Franson Manjali not only did aremarkable

job but, due to his deep competence in cognitive linguistics, as can beevidenced from hisbook Nuclear Semantics (Bahri, 1991), he made many important suggestionswhichenabled me to improve upon the original text. This new version is nowmetaphysicallyoelighter and more completely focused on its scientific substance. Theoecontinentalphilosophical digressions have been almost completely expunged.Supposing this book can have any relevance, I think it is mainly as a precursorof


13/83

the works on topological and dynamical models which have became so widelyaccepted inthe cognitive sciences during the nineties. Thirty years ago, the very idea thatphysicomathematicalmodels of this type could be developed for explaining perceptual, linguistic,and semiotic structures was not clearly understood. It was taken for grantedthat the onlyavailable formalization in the cognitive science fields had to be, for principledreasons, ofa logico-algebraic and combinatorial type. In this context Ren Thomsseminal idea of analternative morphodynamical paradigm triggered off a true scientific revolution.It settledthe basis for a dynamical approach to higher level cognitive tasks such ascategorizationand syntax.As far as I know, it was Christopher Zeeman who introduced the first dynamical

approach for explaining the links between neuroscience and psychology. In hisseminal1965 article Topology of the Brain, he introduced the key idea that brain activitymust bemodeled by dynamical systems on high dimensional configuration spaces ofneuralactivities. Mental states were then identified with attractors of these dynamics,theirM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 19 of 99

content with the topological structure of the attractors, and the flow ofconsciousness witha oeslow temporal evolution of the neural dynamics. Consequently, thestrategy forexplaining mental phenomena was to use the mathematical theory of dynamicalsystems(global analysis) especially theorems concerning the general structure ofthe attractorsand their bifurcations for drawing empirical conclusions from this dynamicalperspective.

This strategy was very clearly outlined by Zeeman in his 1976 article, oeBrainmodelling :oeWhat is needed for the brain is a medium-scale theory. (...) The smallscale

theory is neurology : the static structure is described by thehistology of neurons and synapses, etc., and the dynamic behaviour isconcerned with the electrochemical activity of the nerve impulse, etc.Meanwhile the large-scale theory is psychology : the static structure isdescribed by instinct and memory, and the dynamic behaviour isconcerned with thinking, feeling, observing, experiencing, responding,remembering, deciding, acting, etc. It is difficult to bridge the gapPage 1810


14/83

between large and small without some medium-scale link. Of course thestatic structure of the medium-scale is fairly well understood, and isdescribed by the anatomy of the main organs and main pathways in thebrain. (...) But what is strikingly absent is any well developed theory ofthe dynamic behaviour of the medium-scale.Question : what type of mathematics therefore should we use todescribe the medium-scale dynamic ? Answer : the most obviousfeature of the brain is its oscillatory nature, and so the most obvioustool to use is differential dynamical systems. In other words for eachorgan O in the brain we model the states ofO by some very highdimensional manifold M and model the activity ofO by a dynamic on M(that is a vector field or flow on M). Moreover since the brain containsseveral hierarchies of strongly connected organs, we should expect tohave to use several hierarchies of strongly coupled dynamics. Such amodel must necessarily remain implicit because it is much too large tomeasure, compute, or even describe quantitatively. Nevertheless suchmodels are amenable in one important aspect, namely their

discontinuities. (Zeeman, 1977: 287)M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 20 of 99

It is precisely using these results of global analysis, bifurcation theory andsingularity theory, that Ren Thom worked out his research program leadingfromphysics to cognitive sciences, including linguistics. His main idea was to usethese toolsfor developing a unified mathematical theory of natural morphologies andcognitivestructures.He showed, first of all, that, insofar as it concerns the system of relations whichlinks up parts within a whole, every structure is reducible to a (self)-organizedand (self)-regulated morphology. But, as we will see in a detailed manner in this book,everymorphology is itself reducible to a system of qualitative discontinuities emergingfrom theunderlying substrate (be it physical, neural, purely geometrical, or evenoesemantic). Thetheoretical problem was therefore to build up dynamical mechanisms whichwere able togenerate, in a structurally stable way, these discontinuities both at the local and

the globallevels.Deep mathematical theorems have made possible a revolutionary strategywhichcan be called dynamical functionalism. Instead of first defining the generatingdynamicsexplicitlyand then deriving from it the observable discontinuities, one firstdescribes the


15/83

observable discontinuities geometrically and then derives from them aminimally complexgenerating dynamics. This minimal explicit dynamics must be conceived of as asimplification of the real implicit generating dynamics. This dynamicalfunctionalism isnot of a classical (e.g. Fodorian) type. Indeed, classical functionalism entails astrictseparation between the cognitive and physical levels, the relation between thetwo being amatter of mere compilation and implementation. This is no longer the case in anemergentist (supervenient) approach. But dynamical functionalism isnevertheless a oetruefunctionalism in the sense that emergent structures share properties ofuniversality whichPage 1911are to a large extent independent of the specific physical properties of their

underlyingsubstrata.Such an explanatory paradigm has been extensively developed during theM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 21 of 99

seventies and the early eighties. In physics, and particularly in macrophysics,morphodynamics has innumerable applications. They concern the mathematicalanalysisof the singularities and discontinuities which emerge at the macro level fromunderlyingmicro-physical mechanisms. Here is a very incomplete list : caustics in optics;phase

transitions, symmetry breaking and critical phenomena; elastic buckling;defaults inordered media; shock waves; singularities of variational problems; dissipativestructures;changes of regimes in hydrodynamics, routes towards turbulence; deterministicchaos;etc. The main import of these mathematical models is to explain how theobservablemorphologies which dominate the phenomenologically experienced world canemergefrom the underlying physics. They bridge the gap between physical objectivity

andcommon-sense realism, a gap which arose in the aftermath of the Galileanrevolution. Inthat sense, morphodynamics can be considered as the pure mathematical wayleading toqualitative physics. More than ten years before the computational (ArtificialIntelligence)approach was introduced, it showed that the informationally relevant and salientfeatures


16/83

of macro-physical processes are constituted by their singularities, theirqualitativediscontinuities and their critical behavior.But one of the most significant achievements of Thoms paradigm concerneditsapplication to cognitive processes such as perception, action and language. Itgave anextraordinary new impulse to traditions such as Gestalt theory, phenomenologyandstructuralism. It was for the first time that, in cognitive and linguistic matters,differentialgeometry could substitute formal logic as the main mathematical tool.But Thom and Zeeman proceeded as mathematicians, not in a oebottom-upmanner, from empirical data first to ad hoc models and, at the end of the line, totheoretical principles, but rather in a oetop-down manner, fromfundamental principles

and mathematical structures to empirical data. The advantage of such astrategy was thattheir perspective was theoretically very well grounded and mathematically verystrong.

Their dynamical functionalism introduced a new level of functional architecturewhichcould operate as a condition of possibility for the implementation of syntacticprocessesinto the brain dynamics. The limits of such an approach were of course the lackof aneffective computational theory to undergird it.

Since the early nineties things have radically changed essentially becausedynamical models such as connectionist ones became computationally effective.One cannow say along with Tim van Gelder, that the dynamical paradigm has becomedominantrelative to the logico-combinatorial one. I think that one of the main challengesof futureresearch will be to synthesize the two paradigms.M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 22 of 99Page 2012

So, basically this book can be read as a pioneering attempt to introducemorphodynamical models in structural linguistics and semiotics.I would like to once again acknowledge my debt to Franson Manjali. The longdiscussions with him on this translation have proved to be a theoreticallybeneficialopportunity.I want also thank my colleagues and friends Per Aage Brandt and WolfgangWildgen for having accepted this text in their Peter Lang series.Indian Institute of Advanced Study


17/83

Shimla, April 1999M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 23 of 99Page 21

INTRODUCTION

1. This work is devoted to a study of the applications of Catastrophe theoreticalmodeling and of the epistemological issues deriving from it. We will be mainlyconcernedwith the fields of structural linguistics and semio-narrative structures. Theinvestigationproceeds at two levels. At the level of modeling we show that the topologicalanddynamical syntax conceived of by Ren Thom allows us to tackle and evenpartially solvesome of the main difficulties encountered in structuralism. 1 At theepistemological level,we examine the relevance ofgeometric notions in the language sciences, and

concludethat they provide a schematization in the sense of a geometrization of themeaningof theoretical concepts of the theoretical categories of structuralism. Weaim thereforeat a constitution of the structural domain. Even though this constitution is notstrictly of aphysical order, to the extent it uses mathematics to reconstruct empiricalphenomena, it isof a physical type.2. From a detailed study of the various structural conceptions, we see that,

whateverM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 24 of 99

the domain considered, we come up with a primitive concept of structure whoseformalcontent has not yet been adequately mathematized.(i) In the domain of biological organization, we have to understand how thefunctionof parts in relation to a whole depends on their interdependentpositions. If astructurecan exist, it is because parts are determined reciprocally through a dynamic

process

which defines theirpositional values. This is what Geoffroy Saint Hilaire alreadycalledthe principle ofconnection.(ii) In the domain of perceptual organization, a similar problem is posed by theexistence ofGestalt structures.(iii) In phonology, the phonemes are conceived of as abstract discriminatingunits


18/83

(types) which are equivalence classes of allophones (tokens). Now, these classesarealso defined by an underlying principle of connection. They are obtained fromthecategorization of audio-acoustic substrata, and are positional values withinphoneticparadigms.(iv) In syntax, the primitive structures are constituted of reciprocally determinedactantial places. 2They also provide, though in a somewhat different way,positionalvalues arising from connections. These connections are semantic, and notformal1 By oestructuralism we mean here the tradition founded by de Saussure and further developpedbyTroubetzkoi, Jakobson, Tesnire, Hjelmslev, Brndal, Lvi-Strauss and Greimas.2 We use here the terms oeactant, oeactantial, oeactantiality in the sense ofTesnire (1959) and Greimas(1966). These key words of European linguistics concern the semantic roles of case grammars.Page 2214relations. They belong to the form of content (in the sense of Hjelmslev). Theyareindependent of lexical features and constrain the grammatical function of theterms theyconnect. They belong to a conceptual syntax, and not a formal one. Theircontent ispurely positional.(v) Finally, in the semiotics of narrative, Greimasian theory employs thephonologicaland the actantial models to explain semantic and syntactic organizations

respectively. ItM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 25 of 99

thus combines two structuralist conceptions, and considers semantics in aparadigmaticway (like phonology). The main problem is therefore to understand the linkagesbetweenthem. The key idea is that of a oeconversion of the semantic paradigmsinto actantial(syntactic) interactions, what is called in structuralist traditions theoeprojection of theparadigmatic axis onto the syntagmatic one.In all these domains, structuralist theory depends crucially and ultimately on theformal contentthat must be ascribed to the category of connection, and thus onthemathematization of the concept of positional value. Only such a schematizationcanrightfully establish a oephysics of structures. But it depends, in turn, onthe invention of a


19/83

geometry of position that can describe and explain the organization, thestability and theclosure of elementary structures as well as the constraints imposed on theircombinatorics.It depends on the construction of a new kind of general dynamics, of an originaloeanalysis situs, which still remains a tremendous challenge.Indeed, as Buffon and Kant had observed, such an analysis situs oeis totallylacking in our mathematical sciences. This oetotal lack has, until now,played the role ofa sort of blind spot in our vision of rationality; it has been anoeepistemological obstacle(in the sense of Bachelard) to the constitution of structural objectivity. It hasmadestructuralist theories to keep swaying between psychological reductionism,idealistvitalism and logical formalism, three positions which are not acceptable exceptdogmatically. 1

In linguistics, the formalist approach remains dominant. Based on the fallaciousevidence, borrowed from logical positivism, that mathematics is a languagewhichprovides the most typical example of syntax / semantics relation, it reducesstructures tomere formal combinations. Thus, it is forced to discard the concrete dynamicaloeorganicity of structures in favour of a system of abstract relationsbetween terms. As thefounders of the Gestalttheorie had remarked, this involves a reification ofconnections,which, by attributing to static terms all that in fact belongs to positional values,

ignore thedynamical nature of structures. As far as structures are concerned, formalizationisopposed to mathematization. Thus, there exists a conflict between the formaltreatment ofstructures and their oemathematical physics. The former is associatedwith a formal logic1 See, sections I.2 and I.3.M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 26 of 99Page 2315of terms and relations while the latter refers to a dynamic topology of placesandconnections.3. Catastrophe Theory offers the first instance of analysis situs of structures. Itremoves, at least in principle, the epistemological obstacle which has until nowpreventedthe constitution of the structural objectivity. We intend to show that thistheoreticalpossibility is also a pratical one.


20/83

4. We will reserve another work for a detailed elaboration of the catastrophistformalization of semio-narrative structures. 1 In the present work, we shall focuson someimportant theoretical questions regarding the above indicated issues.In the first chapter we will present a problematized panorama of various criticalissues of structuralism. We have tried to give the study as much theoreticalopening aspossible. We shall refer to structural biology, Gestalt theory, phenomenologyandtranscendental philosophy. This is necessary in order to trace theoememory of thestructuralist idea and to retrieve all its sharpness and amplitude. In Chapter II,we shalldiscuss in more detail the two basic structuralist conceptions, namely

Jakobsonianphonology and structural syntax. This will allow us in Chapter III to revisit thefoundations of Greimass theory of semio-narrative structures.

5. The main part of this oephysics ofmeaning had been developedbetween 1972 and1976. 2 If we have postponed its exposition till now, # it is because we stumbleduponphilosophical difficulties concerning the epistemological status of the modelingofstructures as natural phenomena. As we know, structures have beentraditionnallyunderstood in symbolic terms, that is as constituted of formal relations. Asignificant leaphad to be taken to reach the naturalist conception. The main point is the

following. Inphysical sciences, concepts are not only descriptive, but can also betransformed intoalgorithms for reconstructing the diversity of phenomena. If we take thestructures ofmeaning as natural phenomena in a physicalist sense, we need to transformthestructuralist concepts which describe them into algorithms for reconstructingtheirdiversity.M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 27 of 99

1 See Petitot 1992.2 See Petitot 1977b, 1977c, 1979c, 1979d.#That is 1983.Page 2416

This is how we were convinced that a oephysics ofmeaning has to befounded ona mathematical schematization of categories of structuralism. In order to stressthis idea


21/83

we have called our project a oeschematism of structure1.6. Assuming that our work has some interest and some originality, we hope wewillbe able to convince the reader that far from becoming obsolete, structuralism ison thecontrary in the process of becoming a new frontier of science. We now have thepossibility of extending the physical rationalism into a structural rationalism,mathematically founded, encompassing symbolic and semiotic orders. We nowhave thepossibility, by extending natural ontology, of naturalizing meaning without anylongerhaving to sway between its symbolic reification and its existential experience.

Torre Pellice, August 1983M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 28 of 991 Our notion of schematism is not exactly that of Kants transcendental schema. It concernsschematismas a oeconstruction procedure for concepts.Page 25

CHAPTER IPROBLEMATIC ASPECTS AND KEYISSUES OFSTRUCTURALISMIn this first chapter we shall describe methodically, though not exhaustively,someof the most significant aspects of dynamical structuralism (Section 1 and 2). Thiswill leadM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 29 of 99

us to an inquiry into the conditions of possibility of mathematizing structures(Sec.3.1).As Gilles Deleuze has shown in a essay that we will discuss (Sec.3.3), thefoundations ofstructuralism are topological and not logical (Sec.3.2). Until now, theabsence ofsuch foundations have been obfuscated by speculative interpretations becauseof the lackof any adequate geometry (Sec. 4). In conclusion, we shall briefly summarizetheprinciples of Catastrophe Theory (Sec. 5).

1 . UNDERSTANDING SToeRUCTURE Depending on the domain considered, the concept of structure can have quitedifferent contents and epistemological values. In the case of a mechanicaldevice, aconstruction, or a work of art, we can generally describe the structure in termsof itsdesign. In the case of physico-chemical systems (e.g. crystals, macromolecules,etc.) we


22/83

can also derive the structure from the interactions between its components. Forexample,the progress made in molecular biology and in microbiology has resulted in adecisiveadvance in the comprehension of the structure (stereo-chemical composition) ofDNAchains, proteins, enzymes, membranes, etc. There are of course considerableexperimental difficulties. Their solution requires highly sophisticatedtechnologies ofobservation and reconstruction. But, in principle, even if one does not fullyconfuse thestructure with the observed morphology, even if one considers the former as theorganizing principle underlying the latter, the ontological status of thereconstructedstructures is not at all problematical.On the contrary, in other domains, such as naturalist biology (taxonomic,anatomical, morphogenetic), perception, anthropology or semio-linguistics, one

encounters non-material supervenientstructures, abstractforms of organizationwhichare not directly reducible to systems of components in interaction. This simplefact raisesconsiderable theoretical problems to the extent that we cannot any longer,without furtherinquiry, regard the structures as empirically given phenomena and objects ofexperienceendowed with a predefined ontological status. The very objectivityof structuresmustPage 26M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PM

http://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 30 of 9918then be constituted as such and that is why, in all these cases, a deeperreflection leadsus:(i) to promote the organizational concept of structure to the level of afundamentalcategory of scientific thinking,(ii) to investigate its objective value, and(iii) to seek ways to mathematize its categorial content.In naturalist and descriptive biological sciences, as in social sciences,

structuralismrepresents a rationalistattitude, emphasizing the role of theory andformalization. Itspoint of view is opposed to empiricist reductionism as well as historicistevolutionism.

The shifts from atomistic psychology to Gestalttheorie, from comparative andhistoricallinguistics of the oeneo-grammarians to Saussurian structural linguistics inEurope, or


23/83

from oebehaviorist linguistics to generative or cognitive grammars in theUnited States,from biographical and socio-psychological literary criticism to structuralistcriticism, etc.,are trends in the direction of a general philosophy of systems conceived as rule-governedwholes. In this sense, the horizon of structuralism is that of a theoreticaldescription offormal dependence relations which oeorganically connect the parts in awhole.In so far as it is the ideal form of the organization of a substance, a structure isnota sensible phenomenon. Though it is invisible as such, its substantialrealizations and itseffects are observable and can be subjected to well-defined experimentalprocedures. Inthis sense, every structure is a theoretical object and not a fact. If we want

to avoidnaive idealism, we have to constitute it as an object of experience, as a formemergingfrom the organization of the substrata where it is implemented. Thus, weencounter here aoefoundational aporia, to use Ren Thom's expression. As GillesDeleuze claims, astructure is oereal without being actual, ideal without being abstract; it isa pure oevirtualityof coexistence which pre-exists being; it is oeembodied (implemented)in its substratum,

but is never actualized as such. 1The sensible expression of a structure isalways anegation of its ideal essence. That is why, as Krzysztof Pomian observes, allstructuralapproaches substitute the initial observed objects such as language, naturalforms, etc.with pairs of objects whose ontological statuses are different:oeparole and langue (Saussure), allophones and phonemes (Jakobson,

Trubetzkoi), substance and form (Hjelmslev), systems of kinship andelementary structures of kinship (Lvi-Strauss), performance andcompetence (Chomsky), empirical morphologies and their underlyingdynamics (Thom), etc. Each of the first terms of these pairs (which onemight call oerealizations) are accessible to sensory experience, or toM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 31 of 99

observation, and their reality consists in this. Each of the second terms,the structures () cannot by definition be perceived or observed; wegrant them a reality on the basis of a demonstration, more or lessrigorous depending upon the case. The relations between realization andstructures are variable, but it is always the latter which render the


24/83

1 Deleuze, 1973: 313. We will discuss Deleuze's views in section 3.3.Page 2719former stable and intelligible. As a result, structures are defined as thesets of rational and interdependent relations, whose reality isdemonstrated, whose description is provided by a theory, and which

are realized by a visible or observable object whose stability andintelligibility are conditioned by them.1Given such a status ideal and non phenomenal in the classical sense ,structures are thus ontologically ambiguous. As Umberto Eco asked:oeIs the structure an object, in such as it is structured, or rather the set ofrelations which structures the object, and can be abstracted from it ? 2In fact, as eidos, a structure is not detachable from the substance where it isactualized. 3But must we consider it as given or asposited? In the first case, one will tend todevelopan ontological (realist) conception of structures while in the second, anepistemological(nominalist) conception.Currently, the epistemological interpretation of the category of structure isdominant. It reduces structure to an operational concept whose reality is notontologicalbut only methodological. However, it should be stressed that all the majorstructuralists(Saussure, Jakobson, Tesnire, Hjelmslev, Piaget, Lvi-Strauss, Chomsky,Greimas,and lastly, Thom) have been or are oerealists, even if they dontengage in a philosophicalquarrel.

In fact, from an epistemological, methodological and oenominalistperspective, theconcept of structure can only be a descriptive concept, indeed empiricallybased, butepiphenomenal and devoid of any objective value of its own. Thoughoperational, it isM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 32 of 99

nothing more than a theoretical construct, an artefact, and cannot by itself be agenuinescientific notion. In particular, it cannot contribute to the mathematization ofphenomena.On the other hand, from a oerealist perspective, it is a concept, thoughinitiallyproblematic, acquiring beyond its empirical validity, an objective value and aconstitutiverole. Via the schematization of its categorial content, it becomes a source ofalgorithms forreconstructing specific classes of phenomena.


25/83

If we wish to subject structuralism to a systematic historical oespectralanalysis,we must analyze at least the following trends.(i) The dynamical structuralism of biological origin, which, starting in Germanphilosophy with the Naturphilosophie and Goethe's Morphologie, hasprogressed, viaDriesch and DArcy Thompson, up to Waddington's concepts ofoemorphogenetic fieldand oechreode. This dynamical structuralism is centered on the problemofmorphogenesis.1 Pomian, 1981: 758.2 Eco, 1968.3 Cf. ibid.Page 2820(ii) The phenomenological and gestaltist structuralism which began early thiscentury

on the basis of Brentanos works with Stumpf, Meinong, Ehrenfels, Husserl,Khler,Koffka, Wertheimer, etc.(iii) The linguistic structuralism resulting from Saussure's oeepistemologicalbreakthrough. As we already stressed, it has become one of the basicparadigms in socialsciences, be it in Phonology with Jakobson, in Anthropology with Lvi-Strauss,inGeneral Linguistics with Tesnire and Benveniste, or in Semiotics withHjelmslev andGreimas. This structuralism is twofold:

(a) the oerealist phenomenological structuralism of Jakobson whichmaintains closerelations with dynamical structuralism and gestalt theory;(b) the formalist structuralism (oemethodological andoeepistemological) of Hjelmslev,M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 33 of 99

Lvi-Strauss,1 Chomsky and Greimas who conceive of structures asoeaxiomatizedtheoretical objects and solve the question of their ontological status byembedding them ingenetically determined cognitive capacities.

(iv) The epigenetic and cognitive structuralism of Piaget.(v) The oecatastrophist structuralism of Ren Thom, which is a profoundsynthesis ofthe concepts ofmorphogenesis and structure. It is the first approach to havesucceeded inmathematizing structures as theoretical objects.

To get a more complete picture, we must also explain certain generalproblematics


26/83

related to the project of structural rationalism. Of these, at least five appear tobe essential.(i) Experimental methods which provide an access to the structures. We haveseenthat structures are non material and ideal, and cannot be directly observed. Afirst methodof access (advocated by Lvi-Strauss) consists in analyzing thetransformations ofstructures by variational procedures. Indeed, if a structure identifies itself with aglobal,internal, and rule-governed system of relations, then every local variation mustimply aglobal transformation manifesting the structure. A second method (that ofChomsky'snative speaker conceived of as a language automaton) involves the use of thetraditionalpractice of introspection as part of the experimental procedure.

(ii) The relation between structure and function. Ever since the historic debateconfronting Geoffroy Saint Hilaire's principle of connection with Cuvier'sprinciple offunctional correlation, there has been in biology a dialectical relationshipbetween aphysicalist attitude (mechanistic and materialist) endorsing aoemicromerist reductionistconception, supporting active experimentation, rooted in physiology and, to day,ofessentially neo-Darwinian inspiration, and a naturalist attitude, endorsing aholistic vitalist

conception, supporting common sense observation, based on morphogenesis,and of asomewhat Lamarckian inspiration. But this debate is often a bit skewed, for thephenomena of adaptation (and in particular those of adaptive convergence andco-1 Lvi-Strauss conception is more complex. It involves also Jakobsonian and biologicalstructuralisms.Page 2921M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 34 of 99

evolution) show that these two positions are rather complementary, and that itis

impossible to privilege one against the other. 1The real problem is rather toexplain thecomplementarity itself.(iii) The relation between structure and teleology (finality). One of the mainreasons fordisfavoring the concept of structure since long, has been essentially the factthat, as


27/83

regards the systematic organization of parts in a whole, it is a teleological idea.Totransform it into an operative scientific concept, we must oede-finalise it.

This is possibleonly by way of its mathematization.(iv) The formalization of structures. It has become commonplace to say that theconcerted development of general theory of systems, cybernetics, and formalontology ofrelationships has enabled the oeaxiomatization of the concept ofstructure. But it must beemphasized that the mechanistic-formalistic approaches are largely insufficient.As wehave seen, they are obtained only by a symbolic reification of structures. Theycannotaccount for the dynamically self-organized and self-regulated emergent(supervenient)forms. In other words, they do not provide an answer to the critical question of

the formsubstancerelationship.(v) The levels of organization. Structure-function complementarity comes up atalllevels of composition and observation. The central question is to define theobjectivereality of these levels and to understand their correlations.In this first chapter, we will provide a preliminary sketch of these diverse issues.We will not speak of the oeclassical structuralism which forms part of thecomtemporaryscientific culture (Saussures structuralism, Parsonss structural-

functionalism, Harrissand Chomskys structural linguistics, the structural analysis of economicequilibria etc.). 2We prefer rather to focus on:(i) the still largely unsolved theoretical problems concerning structuralism;(ii) its morphological , phenomenological, and gestaltist oeaccursed part;(iii) the oerevolution represented by the catastrophist turn.1 See, Delattre et al., 1973.2 For an introduction to structuralism, see for instance the following works : Almansi, 1970; Bach,1965; Badock, 1975; Barthes, 1966; Bastide, 1962; Benoist, 1975; Benveniste, 1966; R. Boudon, 1968,1973; P. Boudon, 1981; Broekman, 1974, Cassirer, 1945; Chomsky, 1965; 1966, 1968; Damisch, 1973;Delattre, 1971; Eco, 1963; Ehrmann, 1966; Gandillac et al., 1965; Glucksmann, 1974; Greimas, 1966;Guillaume, 1979; Harris, 1951, 1970; Hawkes, 1977; Hnault, 1979, 1983; Hjelmslev, 1968, 1971;

M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 35 of 99Jacob and Francone, 1970; Jakobson, 1971; Jakobson and Lvi-Strauss, 1962; Katz and Fodor, 1964;Laughlin, 1974; Leach, 1976; Lvi-Strauss, 1949, 1958, 1964-1971; Macksey-Donato, 1970;Maranda,1966; Marin, 1977; Piaget, 1968; QS, 1973; Raccani and Eco, 1969; Robey, 1973; Saussure, 1915;Sebeok and Osgood, 1965; Segre et al, 1965, Viet, 1965.Page 30222. M AIN TRENDS IN STRUCTURALISM : A BRIEF REVIEW


28/83

In this section we present a brief historical account on some major perspectivesonthe phenomena of (self-)organization. We are dealing with fields (biology,psychology,phenomenology, anthropology, and semio-linguistics) where the concept ofstructure isnot only a descriptive tool but also a means of going beyond the conflictbetween theobjectivist-reductionist explanations and the idealist-holistic ones (see section1).2.1. The aporia of organization in Kants Critique of the Faculty ofTeleological JudgementI think we can locate the origin of the modern structural problematic inKantstreatement of biological organization in terms offinality(Kant called it oetheinternalfinality of natural ends) in his Critique of the Faculty of Judgement; more

precisely inhis demonstration that the theoretical comprehension of organizationnecessarily requiredtwo complementary principles (two oemaxims of judgement), onereductionist, and theother holistic. Let us briefly trace his arguments. 1(i) Given the a priori structure of possible experience, we cannot admit of anyobjective finality in nature. Objectively speaking, nature is necessarilymechanical. Inother words, reductionism is the only objectively valid thesis.(ii) It is however an empirical fact that there exist in nature oenatural

ends, i.e. thingswhich are oecause and effect of themselves, 2 in short, organized livingbeings. TheM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 36 of 99

fundamental features of the natural biological ends are, according to Kant,morphogenesis, regulation (homeostasis), reproduction, and the adaptiverelationshipwith the environment (external finality).(iii) Now though Kant might have accepted that the progress of physics would,oneday, explain mechanistically some of these features, he made the decisive

remark thatsuch an explication would still, for reasons a priori, be incomplete to the extentthat itwould not account for the contingencyof the form of organized beings. For Kant,thecontingency of form is part of the oespecific character of natural ends.Because it eschews


29/83

the laws of geometry and physics, it can be understood only reflectivelyvia theIdea (andnot the category) of finality.1 See, Petitot, 1982d, for a more detailed account.2 Kant, 1790: 190.Page 31

23(iv) The internal finality is not just organization, but self-organization. In anaturalend, there exists a reciprocal determination between the parts and the whole.

The structureis not that of a mechanism, but the effect of the idea of the whole determiningthesystematic unity of the form and the connection between parts. Theorganization dependstherefore on a oeformative force (bildende Kraft), which not beingexplicablemechanically, is not objective. That is why it is an oeunfathomable quality,

anoeincommensurable abyss where reductionism, though the onlyobjectively valid maxim,should nevertheless be treated along with the holistic concept of finality.(v) The reductionist and holistic maxims of judgment seem to be contradictory.

Theyopen out therefore to a natural oedialectic. But, for Kant, the conflict isnot a trueantinomy for it concerns only maxims, i.e. prescriptions that a subject mustfollow forgaining knowledge. Maxims are only heuristics for the comprehension of

phenomena.There would be an antinomy only if, moving dogmatically from reflective todeterminantM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 37 of 99

judgement, we would use the idea of finality as a constitutive concept, as anobjectivecategory. But, even if it is only heuristic, the rational concept of finality isoeas necessary for the human faculty of judgementas if it were anobjective principle1 .(vi) For Kant, the possibility that a regulative Idea can have the same value as acategorial concept comes essentially from the finite (oediscursive,

oenon-intuitive) natureof our understanding.Since Kant, things seem to have notably changed. But this is quite illusory. Theepistemological obstacle masterly identified in the Critique of the Faculty ofTeleological Judgement namely, the principled impossibility of a physicalexplanation of the phenomena ofmorphogenesis, (self-)organization, andregulation


30/83

is still far from eliminated. Indeed, the advances in reductionist biology(molecularbiology and neo-Darwinism) on the one hand, and in the techniques ofcyberneticsimulation on the other hand, have given us a lead. But we are still far fromunderstanding how stable and self-regulated structures can emerge from aphysicochemicalsubstratum. The difficulty is not so much experimental as theoretical. What welack are concepts, not facts. It is only recently that in the physical (non-biological) caseswe have been able to explain, using the theory of bifurcations of dynamicalsystems, howmaterial media can spontaneously self-organize, either purely temporally(oscillatingchemical reactions) or, spatio-temporally (spatial patterns of Belousov-Zhabotinsky1 Kant, 1790: 218 (our translation).

Page 3224reaction, Bnard's cells, etc.). 1 In this sense, biology still remains, as JeanPiaget hadaffirmed, oethe key to structuralism2 .M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 38 of 99

2.2. Structuralism in BiologyIn biology, the structuralist paradigm is a dynamical perspective that appearedwhenever the idea ofmorphogenesis came up. Here, the concept of structureisinseparable from that of form. Therefore it has always been, until recently, tied

up withthe speculative concept ofentelechywhich goes back to Aristotle. This explainswhy itwas rejected by the anti-Aristotelian reductionists.

The issue began with theprinciple of spatial connection between parts in awhole, introduced by Geoffroy Saint Hilaire, and later taken up by Goethe. In hislongand patient meditations on plant morphogenesis, streching from 1770 until hisdeath in1832, Goethe sought not so much to understand the physico-chemicalmechanismsunderlying the formation of organisms, as to discover the principle by which anorganismis what it appears to be. 3 He quickly came to the conclusion that whatdistinguishes anorganism from a machine is the fact that in the case of an organism, theexternalappearance is governed by an internal principle producing the spatial (external)connections between parts. For Goethe, it was the understanding of thisprinciple which


31/83

constituted the central theoretical problem in Biology. However, thoughreferring to anempirical phenomenon, the concept of connection is, as we see with Kant, only aoenoumenal Idea, and not a oedeterminant concept or category.

Transgressing theargument of Kants third Critique, Goethe put forward the hypothesis thatthere existed aschema for this Idea, which could share infinite concrete variations. Tounderstand theresponse of organisms to stimuli as much internal as external, he seeks todetermine theirconstitutive ideal principle, in other words, their formative laws.Goethe gradually recognized this ideal principle in the spatio-temporal unfoldingof an internal organizing force. According to him, it is this oea priorientelechicprinciple that rules the formation of natural ends. But one of the central resultsof the

Kantian Critique is precisely that a noumenal Idea is, in essence, disconnectedfrom spaceand time. Against Kant, Goethe thought of entelechy as a kind of oeintuitiveconcept.Contrary to physics, where concepts are abstractions relative to the sensibleworld, forhim the concept of structure was a real, concrete and perceptual entity. That iswhy1 See, for instance, Prigogine, 1980.2 Piaget, 1968.3 For this account of Goethe's conception, see Steiner, 1884. I thank Filomena Molder who introducedM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-

nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 39 of 99me to this remarkable work.Page 3325entelechy can be an the intuitive concept and an efficient idea, which byunfolding itselfspatio-temporally brings about morphogenesis.Goethe's answer to the aporia of form in biology was of a speculative nature. Itisone of the sources of vitalism. But nevertheless its epistemological valuecontinues to beretained in contemporary trends of dynamical structuralism. As an example we

can referto the defense of structuralism in biology proposed by B. Goodwin and A.Webster, inline with the ideas of the great embryologist Waddington. 1Goodwin and Webster present a historical and epistemological analysis of theclassical conflict between the structuralist and the neo-Darwinian points of view,the latter


32/83

being the synthesis of the Darwinian evolutionary theory and moleculargenetics. Forthem, structuralism is opposed to neo-Darwinian empiricism, not at the level offacts, butas a rationalist point of view in which a priori concepts, categories, andprinciples governthe explanation of empirical data. The central problems they address are thoseof form andmorphogenesis. They investigate the type of categoriality necessary to maketheseconcepts intelligible. Now, the main point is that, by its very evidence, the neo-Darwinianparadigm obscures the intelligibility of morphological phenomena. It reducesthem to aby-product of evolutionary chance, denying thus any oelaws of form.

This is essentially due to the fact that this paradigm confuses the concept ofcontrol with the category ofcause. The genome controls the form and the

development ofan organism at the phenotype level. By acting on the genome one can thereforealsomanipulate its morphological effects. But this causal efficiency does not entailthat thereare no specific and autonomous constraints for forms. By identifying the geneticcontrolof the phenotype with a determinant cause, the neo-Darwinian approachassumes thatthere is nothing to be explained other than the phenomenon of control itself: as

Jacques

Monod claimed, form is causally reducible to the primary structure of proteins,and all therest is only a matter of thermodynamical processes of self-organization.Neo-Darwinism is a materialist reductionism which privileges functional aspects,M O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 40 of 99

reduces structural connections and positional organization of parts to a merespatialcontiguity, and subordinates the oeinternal finality to an oeexternalfinality, i.e., toadaptation and selection. It reduces structure to genetics. For it, structure ishistorically

given , and has only an evolutionary necessity as the epigenetic expression ofits geneticprogramme.Structural rationalism denounces the inconsistency of making history not onlythecause of evolution, but also that ofstabilityand invariance of species. 2According to its


33/83

view, an organism is not only a genetically controlled system, but also astructure, that is1 See Webster, Goodwin, 1981; Waddington, 1956, 1957. For the opposite, neo-Darwinian point ofview, see, for instance Danchin, 1977.2 On this question, see also Gould, 1977, Gould, Eldrege, 1977.Page 3426

a totality organized by a system of internal relations satisfying someoelaws of form. Therealm of organized beings manifests a certain necessity. The structures areneitherirreducibly diverse, nor the arbitrary result of evolution.

The fundamental tenet of structural rationalism is that the expression of thegenotype into the phenotype cannot be completely understood unless weintroduce somesort ofpositional information controlling cellular differentiation. In organizedbeingsthere would be apositional efficiency, the position selecting metabolic regimes

bytriggering the right genes. It is the understanding of such positional informationandefficiency which constitutes the central theoretical problem of dynamicalstructuralism.In the Waddingtonian theory of morphogenetic fields and oechreodes, themaincharacteristics of structural organizations are the following: 11. dynamical genesis, self-regulation and structural stability;2. equipotentiality: structures are not mere systems of interaction ofcomponents, but

include a reciprocal determination ofplaces (positional values);3. equifinality and homeorhesis (epigenotype according to Waddington):development is itself structurally stable as a process, and its final state islargellyM O RP H O G E NE SIS O F M E A NI N G 01/06/2006 06:53 PMhttp://www.google.com/scholar?hl=en&lr=&client=safari&q=cache:2Nil-nPetitot/ArticlesPS/Petitot_MM.pdf+morphogenesis+of+meaning+petitot Page 41 of 99

independent of its initial state;4. the closure of the elementary structures and the existence of constraints, oroelawsof form;5. oegenerativity of forms and the production of complex structures froma closed set

of elementary ones.All these concepts are categories governing morphological phenomena. Theircategoriality (which as we shall see later is more oelinguistic thanphysical) determines thetype of theory we need to render intelligible the morphological and dynamicalconcept ofstructure. We see that the main problem is to give them an objective value.2.3. Gestalt theory and phenomenology

8/3/20

m o Rp h o g e Ne Sis o f m e a Ni n g 01

Documents