Navigazione – Mappa del sito

HomeNumeri76Section one | PhilosophyThe Notion of System in the Work ...

Section one | Philosophy

The Notion of System in the Work of Umberto Eco: Summa, Structure, Code, Encyclopaedia and Rhizome

Claudio Paolucci
p. 39-60


System is a key word for semiotics and linguistics and is a key word also in Umberto Eco’s thought and philosophy. However, Eco always tries to find new words in order to express in a proper way his own idea of system. These new words are also new ways of thinking and rethinking the very core of his own philosophy and semiotics, which remains somehow stable during the years. Through these five words – summa, structure, code, encyclopaedia and rhizome – this paper aims at outlining an image of the evolution of Umberto Eco’s thought and philosophy.

Torna su

Termini di indicizzazione

Torna su

Testo integrale

1. A Philosophical Heritage

  • 1 The Aesthetics of Thomas Aquinas was originally written in 1956. The foreword was written in 1988.
  • 2 As it is well known, Peirce laid the foundations of his pragmaticism on a semiotic theory which gav (...)

1There is a type of short circuit between the foreword of The Aesthetics of Thomas Aquinas and the final chapter of Semiotics and the Philosophy of Language. On the one side, in the foreword of The Aesthetics of Thomas Aquinas (1988),1 Umberto Eco emphasised two important issues concerning the practice of philosophy: i) being the production of a structural model that aims to temporarily make the reality intelligible, a system must have a contradiction that can undermine its validity; ii) the philosopher does not have the task of eliminating the contradictions of a system, rather it must produce them where they do not appear. On the other side, in the finale of Semiotics and the Philosophy of Language, originally published in 1984, Eco identified two aims of philosophy, which referred to Peirce’s theory and his combination of semiotics and pragmatism:2 i) the rejection of something absolutely unknowable and inexpressible; ii) the rejection of any unstructured explanation, such as, for instance, intuition. Eco positioned the cornerstones of his philosophy under the aegis of the notion of a “code”, which in his mind represented the symbol of a semiotic and structuralist koiné – which had always searched for the rules that make it possible to make a reality that requires an explanation intelligible. And these final paragraphs, which bring the book to a close, represent, in my opinion, a veritable summa of Eco’s philosophical heritage.

  • 3 SFL: 300-301.

This metaphorical reference to unity concludes the investigation [...]: viewing the life of culture as a fabric of codes and as a continuous call from code to code has involved searching, in some form or manner, for the rules regulating semiosis based activity. Even when the rules were further simplified, it was important to continue to search for them. The battle for the code was a battle against the ineffable. If there is a rule, there is an institution and there is society, and therefore there is a mechanism that can be built and unbuilt. Hence the speaking of code has meant viewing culture as a matter of regulated interaction, art, language, artefacts, the perception itself as phenomena of collective interaction governed by explicit laws. The life of culture has no longer been viewed as a free creation, a product and object of mystical intuitions, a place of the ineffable, pure emanation of creative energy, a theatre of a Dionysian representation upheld by forces that precede it and something on which the analysis has no hold. The life of culture is the life of texts supported by intertextual laws where every “already told” acts as a possible rule. The already told constitutes the treasure of the encyclopaedia.
It was argued that, with the notion of a code, even where phenomena that are largely unknown occur, there is in principle no unknowable, because something continues to be the object of investigation, and it is the system of the rules, however profound, however intertwined according to a web or also a labyrinth model, and to what extent they can be labile, transient, superficial, dependent on contexts and circumstances.
From this view point, the emphasis and enthusiasm (and also haste) with which post-structuralism has attempted to do justice to the codes and their systems, replacing the rule with the vortex, the béance, the pure difference, the derivation, the possibility of a deconstruction subtracted from any form of control, should not be greeted with excessive enthusiasm. It is not actually a step forward, but a return to the revelry of ineffability. […] We must not forego the energy and enthusiasm with which, from the middle of the century onwards, we took action in order to explain semiotic laws - and therefore human behaviour.3

  • 4 NR: 345.

2The two points of the foreword of The Aesthetics of Thomas Aquinas (written in 1988) logically follow on from the two points of the final chapter of Semiotics and the Philosophy of Language (written in 1984) and depend on the same. It is precisely because we can not and must not accept that something is in principle unknowable, that we must search for and produce structural models that aim to temporarily make the reality intelligible. And precisely because we can not and must not accept intuition or the Dionysian representation upheld by forces on which the analysis has no hold, that we must think that these models can reproduce the contradictory nature of human behaviour. So we must «produce contradictions when they do not appear», we must «put them on stage», as Eco himself went on to say with an expression we will have to return to. It was, after all, the finale of The name of the rose: «And it must also be because, at a time when as philosopher, I doubt the world has an order, I am consoled to discover, if not an order, at least a series of connections in small areas of the world’s affairs».4

3Eco’s theory aims therefore to give an account of the structure of these «series of connections in small areas of the world’s affairs». Eco will never cease to search for rules, laws and the most adequate representation to be able to express them. In other words, Eco will never cease to reconstruct their system. Delving more deeply, one can identify the cornerstone of Eco’s theory by studying the evolution of the idea of system, intended as a «series of structural connections» that «aims to temporarily make the reality intelligible». This work is based precisely on this objective.

2. Five Words for “System”

4During his philosophical adventure, whether it involved aesthetic reflection, semiotics, novels or philosophy, Eco has processed at least five different system concepts: the summa, the structure, the code, the encyclopaedia and the rhizome. The transition phases between these five different ideas allow us to reconstruct the evolution of Eco’s theory, from the first juvenile works on aesthetics through to the final essays on semiotics and novels. Each of these notions does, in actual fact, represent an evolution if compared to the previous one, but it is not a matter of progress or superseding, but rather of an ongoing refinement of a theoretical model that represents, as we have seen, the very meaning of a theory based on Peirce’s principles of anti-intuitionism and the rejection of the unknowable. Some of these notions will in fact be present together, such as the summa and the rhizome within the idea of an encyclopaedia, developed in the mature phase of his theory. Others, apparently abandoned or not fully developed, will return later on in most unexpected places.

  • 5 See Paolucci 2017a, chapter 2.

5In order to understand all of the above, it is worth starting from what Eco writes about an author he has often referred to when talking about himself, concealing his pathway behind that of James Joyce.5

  • 6 PJ: 1044.

In Joyce’s works, the search for a work of art that proposed itself as the equivalent of the world, always moved in one direction only: from the ordered universe of the Summa, which had been proposed to him during his childhood and adolescence, to the universe which unfolds in Finnegans Wake, an open, ever-expanding and proliferating universe, which however must have a module of order, a rule of reading, an equation that defines it and, finally, a form.6

  • 7 In The Role of the Reader, in a translation of the paragraph on the “semantics of metaphor” origina (...)

6This passage «from the ordered universe of the summa» to «an open, ever-expanding and proliferating universe» also characterises the philosophical pathway of Umberto Eco, which starts from a series of works on Thomas Aquinas and medieval aesthetics and arrives at the idea of an encyclopaedia, conceived as an «open and ever-expanding universe» modelled on the structure of Finnegans Wake (see PJ and AL).7 Using the semiotic theory, Eco will never cease to search for «the rule of reading» and «the module of order» of this universe: the “form”, if one wishes to use the terminology of his master Luigi Pareyson.

  • 8 PJ: 1043-1044.

This vision of the universe as a whole – and consequently of its particular forms, in real life and in art – that can receive a single incontrovertible definition in which everything finds its place and reason – has encountered its highest expression in the great medieval summae. Modern culture developed in reaction to this hierarchical view of the universe, but despite opposing it, it has never been able to subtract itself entirely from its charm, from the majestic comfort of an ordered form in which everything is justified. We can say that the history of modern culture was nothing more than the continued opposition between the need for order and the need to identify a changeable form in the world, open to adventure, full of possibilities; but every time an attempt was made to define this new condition within the universe in which we move, we found ourselves holding the formulas of the classical order in our hands, despite being well disguised.8

7Although his own train of thought commences with the liberation from the Thomistic idea of summa researched in his graduation thesis, Eco will never cease to love or to use small summae, ordered totalities of intelligible relationships in which «everything finds its place and reason». This is what his final books on history of ideas were like: History of beauty (SB), History of ugliness (SBR), Vertigo of the List (VL), History of places and legendary lands (STL), Philosophy and its stories (FSS) and all the CD rom encyclopaedias edited and supervised over the years (see, for instance, the Encyclomedia project, 1993-1999). These are small ordered cosmoses of local portions of the Encyclopaedia, used in order to pose the problems under analysis and to rediscover that Thomistic taste for ordered totalities, basically virgins to the taste of Adventure, which see their most significant expression in the great medieval summae.9 I remember Eco was exceptionally pleased after reading a review by Gianni Vattimo, in which he said that, at the end of the day, Eco had never embraced “weak thought” theory because he had always remained a Thomist supporter.10 This in his own way was true, Eco has always maintained a taste for ordered totalities, for classification and hierarchy and for a strong arborescent scaffold that supports the theory, where all parts find their place and their reason.

3. From Summa to Structure

  • 11 For a more detailed and broader overview of the connections between Eco’s semiotics and mediaeval p (...)

8The step from the idea of summa to that of structure, in the sense of structuralism, is very short. Since the 1960s, Eco immediately recognised the similarities between certain ideas of the newborn structural linguistics, which were soon to become the new koiné of human sciences, and Thomistic theories.11 Through the mediation of Pareyson and his theory on form, Eco approached structuralism rather naturally, since, in his opinion, it shared some constitutive principles with the scholastic mindset.

  • 12 SPM: 1245.

Structural re-reading of a medieval text is by no means a random choice. Structuralism finds numerous ascendants in the scholastic forma mentis: the structuralist claim to the interdisciplinary matters, to a universal logic, to the reduction of all human sciences to a leading science (which for structuralism is linguistics), of which the other are all ancillae is decidedly scholastic. Likewise, scholastic thought has two characteristics that relate it to the structural theory: it proceeds using binary partitions (true-false, sic et non, dual structure of the quaestio etc.) and is a synchronic theory. Synchrony for the medieval theory is not a choice of method: it necessarily stems from its metaphysical premises.12

9Once again, this is not a one-way route either. As this passage testifies, Eco comes back from structuralism to a Thomistic scholastic theory, for instance in the republication of his 1970 graduation thesis on the problem of aesthetics, enhanced by a whole series of paragraphs on the relationship between summa and structure and between scholasticism and structuralism. It is no coincidence that, rather late in life, Eco went on to close his work of over 1,300 pages of the Writings on Medieval Thought, by republishing these exact parts, with the most significant title of “Scholasticism and structuralism”.

  • 13 Ibidem.
  • 14 On this point, see the famous chapter of Jakobson’s Writings in general linguistics.
  • 15 On these topics, see the extended discussion in Strutturalismo e interpretazione (Paolucci 2010), c (...)

10According to Eco, in fact, «Thomistic based aesthetics is an aesthetic of form as a structure decomposable into simple elements, united by a relational relationship. And it is [...] a synchronic aesthetics».13 The concept that Eco had of a structure was already very clear (see SA, SFL): it is a relational set of places and positions, capable of describing the synchronic form of the system through binary relationships of opposition. Eco undoubtedly had in mind the successes of the structural phonology of Trubeckoj and Jakobson, who had been able to describe the production of the totality of the sounds of languages (substance of expression) starting from a finite number of relations between phonemes (form of expression), all referable to two major general binary oppositions: A VS B (qualitative opposition) and A VS non-A (privative opposition).14 Indeed, the relationship between bilabial (A) and labiodental (B) is a relation of the first type, in which the opposition between phonemes is defined by the presence of a trait (union of the two lips VS union of the lip and teeth) based on an axis common to both (labiality). On the contrary, the relationship between vocalic (A) and non-vocalic (non-A) is a relation of the second type, where a phoneme can be defined either by the presence of a trait (A) or by its absence (non-A): the trait of which that phoneme is lacking (from here: “privative opposition”). Such oppositions are exclusive, i.e. the presence of a term of the opposition excludes the simultaneous presence of the opposite term, and are based on the functioning of the human phonatory apparatus which, due to biological limits, is not capable of producing more than one sound at a time (hence the “binarism”).15 This idea of structure, formed by binary relations between terms that present a relational identity, is used to describe the form of expression, i.e. a synchronic set of places and positions that precedes the use that speakers make of it. This is the idea of “form” that comes from linguistics: not a platonic structure, but a local set of differential relations that constitute a system of values.

11It is in fact the intrinsic limits of this structure concept, elaborated to respond to a particular and very precise problem – that of the expression plane system of the natural human languages – which will drive Eco to the second transition, that from structure to code. This second transition follows the criticisms that Eco himself, in La struttura assente (1968), already applied to those structuralists that had used the structural methodology for ontological, metaphysical and psychoanalytic purposes. According to Eco, the objective of the inquiry is in fact to produce a structural model that aims to temporarily make the reality intelligible. And, as far as the system of meaning is concerned, the idea of structure simply does not fit for this purpose.

4. From Structure to Code

  • 16 See Violi 1997.

12The second transition, from the structure to the code, is in fact the transition that will take Eco from aesthetics to semiotics, from Il problema estetico in Tommaso and The Open Work, to Le forme del contenuto and A Theory of Semiotics. Indeed, Eco’s main theoretical problem in a book like Le forme del contenuto (1971) is whether it is possible to describe the organization of the systems of meaning (form of content) by means of the same analytical techniques that had performed so well when adopted to analyse the signifier (form of expression). Eco had the precedent of Structural semantics by Greimas (1966), which attempted to construct an “in langue” semantic theory using the techniques of structural phonology, following the postulate of an unproven isomorphism between the language plans. Another attempt in this sense, which Eco himself dealt with extensively, was provided by Louis Hjelmslev, in his theory of the forms of content. Now, both attempts were in fact failures.16 And Eco himself (FC, TSG, SFL) demonstrated this in detail in his books written at the time. In fact, the content plan has a different form compared to the expression plan. Greimas himself, for instance, was perfectly aware of it, to the extent where he constructed his next theory of the generative trajectory of meaning starting from the failure of his Structural semantics. The expression plane is not isomorphic to the content plane: the human phonatory apparatus does not function as meaning does and it is not possible to account and give reasons of the meanings of the language starting from binary oppositions of exclusive type (A vs B, A vs non-A), as it had been done with the sounds of the language.

13Hence the different responses given by Eco and Greimas, and the divide of the theoretical exploits of the two most important semiotic theories of the second half of the twentieth century, the interpretative theory (Eco) and the generative theory (Greimas).

  • 17 See Greimas, Courtés 1979, “Text”, see also Marrone 2010.

14According to Greimas, although it is certainly not possible to build a theory of content forms starting from the theoretical models and the analysis techniques that the structuralists had used for the expression plane, it is necessary to refrain from throwing everything away and maintain the very same structural methodology by applying it to «semantic micro-universes», which Greimas (1970, 1983) later referred to as “texts”.17 In fact, the enormous success of Greimassian semiotics derives from its initial setback: originally designed to construct a semantic theory for natural language (Greimas was a lexicographer), it only proves to be valid within more local and less complex semiotic parts (texts).

15Eco’s response is completely different, even if it follows a similar path of awareness. Since it is not possible to construct a theory of the forms of content using the same structuralist methodologies that had achieved highly significant results at a forms of expression theory level, it is necessary to change the explanatory model, without foregoing the construction of a forms of content theory, which is also the title of the book written by Eco in 1971, which went on to constitute the cornerstone of A Theory of Semiotics. This led to the transition from the structure to the code, more suitable, according to Eco, to describe systems like the semantic one.

  • 18 See Deleuze 1969.

16According to Eco, a structure is not a code, but a monoplanar system of positions and oppositions, in which the terms are reciprocally determined. Eco calls these structures S-Codes (where “S” stands for “system” or “structure”) and thinks that S-Codes are: i) the structure of parental relations studied by Lévi-Strauss, ii) the structure of the phonological system studied by Jakobson, iii) the structure of the system of content studied by Greimas (1966), as well as all the topological and relational systems that represent the constitutive trait “from which structuralism is recognised”.18

  • 19 SFL: 265.

The elements of a phonological system have no meaning, they do not correspond to anything, they cannot be correlated to any content. The distinctive features that constitute and reciprocally characterise the phonemes, are part of a pure system of positions and oppositions, a structure. The absence or presence of one or more traits (expressible and calculable in binary terms) distinguishes one phoneme from another. A phonological system is supported by a (systematic) rule, but this rule is not a code.19

17So why then, and precisely in the years in which he is most influenced by structuralism, does Eco believe that the idea of structure is not enough to represent the suitable model of the system he is trying to identify? It is now possible to identify a core point, which would go on to become a flagship aspect of Eco’s theories, as well as of semiotics itself: with s-codes (structures) one cannot make false statements and process lies about the outside world.

  • 20 SF: 282.

It has been said that s-codes are systems of units that can be defined by their mutual position and that none of these entities is related to a content. Given they are not related to any content, none of these entities can be used for reference operations. In other words, a code can be used to make false statements about a condition in the world, like when a secret agent digits a message in code to deceive the enemy, a librarian identifies a book that does not exist or a natural language user says there are six apples on the table while in reality there are seven. With an s-code however, one cannot designate conditions in the world and consequently false statements cannot be made: one can only make incorrect statements, i.e. statements that breach the internal rules of the s-code in question.20

  • 21 TSG: 13.

18It is the semiotic threshold of the falsehood that drives Eco to replace the idea of structure with the idea of code, where the code is actually something that allows the falsehood and semiotics is the discipline that studies “everything that can be used in order to lie”.21 This is a good example of the internal hierarchy of operations and traits that, according to Eco, allows us to identify when semiotics is present and when it is not. In fact, even a simple structure (s-code) allows interpretations, expectations and inference systems, but this does not suffice in order to speak of semiotics:

  • 22 SFL: 281.

A monoplanar system can allow signification processes, not because it provides correlations, but because it stimulates inferences, i.e. interpretations. A certain position on the board may appear to be wrong, risky or promising depending on the next move in the game, and it is the game (as a concrete actualisation of the rules of the game) which establishes, among the various possible positions on the board, a hierarchy of preferences between positions that “suggest” good development opportunities and positions that “suggest” situations of danger for at least one of the two contenders.22

19It is not therefore the presence of an inferential structure or that of an interpretation system that traces the distinction between a semiotic system and a non-semiotic system: the threshold is higher and concerns the ability to lie. For semiosis to appear, there must be inferences and interpretations, but these are only necessary conditions, not sufficient conditions. Not surprisingly, they are also present within the s-codes. But it is only with a proper code that something can be used to lie. So, what is a “code” according to Eco?

  • 23 SFL: chapter 5.

20It is a fundamental theoretical object for Eco’s own theory, which will then be transformed into that of an “encyclopaedia”, but which, even after its “transformation”, will be elected by the same Eco as a symbol of his philosophical enterprise, as well as of that of the authors Eco drew inspiration from.23 Not surprisingly, in A Theory of Semiotics, the most substantial part of the book is dedicated to the theories of codes, where the “code” is an object with its own identity, something which, back in 1975, was not yet superimposable to that of the encyclopaedia.

  • 24 TSG: 56, 74, 77, 179.

We will use the term code in the proper sense to refer to the rule that links the elements of an s-code with the elements of another s-code. […] The s-codes are in actual fact systems or structures. […] When a code links the elements of a carrier system with the elements of a conveyed system, the former becomes the expression of the latter which, in turn, becomes the content of the former. […] Therefore, a code establishes the correlation of an expression plane (in its purely formal and systematic aspect) with a content plane (in its purely formal and systematic aspect). […] Semiotics gives us a glimpse of a sort of molecular landscape where what everyday perception presents to us as closed forms, are actually the transitory result of chemical aggregations and the so-called “things” are the superficial appearance of an underlying network of more microscopic units. […] What has been referred to as “the code” is a complex network of sub-codes […]. It should be called hypercode (as when we speak of hypercubes) which unites various sub-codes, some of which are strong and stable, others weaker and more transient.24

  • 25 See Paolucci 2010, chapter 4.
  • 26 SFL, chapter 2.
  • 27 Barthes 1970: 24.

21According to Eco, it is the code that establishes the semiotic correlation and defines something as an expression and something else as a content: since A Theory of Semiotics, the code defines the form of semiotics, not that of semantics (form of content). In fact, not only does each culture organise a content system, but it also inscribes a part of itself that is able to express this very same organization.25 The cultural system has the task of organising some of its own parts in order to expresses itself, manifesting it by means of a semiotic function. The expression plane represents these parts and the code keeps track of this organization. Starting from what we have said, the transition from the code to the encyclopaedia is exceptionally short: if the cultural system organizes a system of values of content and inscribes a part of itself that is capable of expressing it, the encyclopaedia, as the “library of libraries”, keeps track of this organization, which becomes «the registered set of all interpretations»26. We will return to the organization of the encyclopaedia in the paragraph below. However, it is actually because of this nature that, in Semiotics and the philosophy of language, Eco, returning to the idea of a code, quotes a passage from S/Z by Roland Barthes, that states that «the code is not a list, a paradigm that must be rebuilt at all costs. The code is a perspective of quotations, a mirage of structures... they are simply glimpsing of that something that has always been already read, seen, done, lived: the code is the furrow of this ‘already’».27

22For Eco (SFL, AL), «the furrow of this already» is called the Encyclopaedia.

5. From Code to Encyclopaedia

23Being simultaneously the combination of the “already told” and the correlation code between expression and content, the encyclopaedia plays a central role in Eco’s theory, representing the most advanced configuration of his idea of a system. Its main characteristic, which allows one, for instance to distinguish the semiotic approach from that of the analytical philosophy of language, is the fact that it is a semantic system that provides pragmatic instructions. For the encyclopaedic type of semiotic semantics developed by Eco, there can be no distinction between semantics and pragmatics and pragmatics must be considered to be a constitutive part of semantics.

  • 28 SFL: 295.

All it takes is for two systems to become intertwined […] and the so-called code (already a system of multiple codes) is no longer simply an apparatus that provides equivalences, but rather a machine that provides instructions for manoeuvring different systems of equivalences in different contexts or circumstances. At this point we are already in the pragmatic dimension: but if the instructions for manoeuvring within the pragmatic dimension are somehow foreseen and provided by the code, well this means that this code (capable of integrating its elementary semantics with pragmatics) has already taken on the appearance of an encyclopaedia, albeit at a minimum level.28

24Identifying the form of the Encyclopaedia is a fundamental task for Eco. This is why he spends so much time eliminating false leads, asking himself, first and foremost, what Encyclopaedia is not. And it is a hecatomb of all the places where his heart beats.

  • 29 See Paolucci 2016, 2017a.
  • 30 SS: 359-360.

25The first solution, that he instantly discards, is that of his youthful love: Thomas Aquinas. One cannot conceive the Encyclopaedia in the image of scholastic and medieval Summae, as «an ordered totality of intelligible relations» (PET, SPM). The Encyclopaedia does not have the form of Order, or does so only locally (SFL, SS). At a global level, it may have the form of Adventure (SS, AL),29 which is precisely the counterpart of «the place where everything finds its place and reason» which «has encountered its highest expression in the great medieval summae ». For the very same reason, Eco also discards the solutions of the much-loved enlightenment Encyclopaedie. Just as one can no longer create a Summa of knowledge, like Thomas was attempting, it is not possible to create a tree of the sciences, as D’Alembert and Diderot wished, trying to replace the «labyrinth» of the «general system of sciences and arts», with an «encyclopaedic tree in which one wants to represent it».30 As the title of his last book on these topics clearly says, Eco deemed it necessary to pass «from the tree to the labyrinth» (AL).

  • 31 SS: 334-9.

26The third solution that Eco discards is precisely that of his «new love» back in the seventies: structuralism. The Encyclopaedia cannot be a structure made of a set of differential relationships, like those described by Lévi-Strauss in the Elementary structures of kinship or by Greimas in his «elementary structure of signification». The Encyclopaedia is a complex and contradictory organism, that should be studied at a systems theory level and has nothing to do with elementary structures of relationships at a metalinguistic level.31 The Encyclopaedia should not be ordered, it should be rendered in the complexity and contradictoriness of its form, which is that of Adventure.

27Not a Summa, nor a tree, nor a structure, Eco sees the Encyclopaedia as the form of «an open universe, in continuous expansion and proliferation». The «module of order», «the equation that defines it» is therefore for Eco the rhizome (SFL, SS, AL, LLP).

  • 32 LLP: 44.

In several books I have already developed the notion of an encyclopaedia, conceived as a galaxy of knowledge that does not embrace the form of a tree, but rather that of a network. An enlightening idea had already come to me in relation to a model by Ross Quillan that I developed in A Theory of Semiotics and, later on, from the idea of a rhizome suggested by Deleuze and Guattari.32

28Deleuze and Guattari (1980) introduced the notion of «rhizome» starting from the theory of complexity and the theory of systems with no centre by Rosenstiehl and Petitot (1977). The rhizome does not have a hierarchical structure, nor does it have a centre of organization, although locally its domains can take on an ordered and arborescent form. However, at a global level, the rhizome is defined as a network in which «any point can be connected to any other» and where the form of the system depends on the paths that cross it. The rhizome, as an encyclopaedic network model, presented some important features, which Eco studied in detail (SS, SFL, AL). According to Eco,

  1. the rhizome, like the Encyclopaedia, is intrinsically contradictory (non-coherent): it admits and incites at the same time “if p then q” and “if p then non-q” (SFL, AL). This characteristic is obvious: if the encyclopaedia is the library of all libraries, which holds together all the interpretations and all the “versions of the world”, then these interpretations and these versions will often contradict each other. The form of the encyclopaedia must therefore be a form which can recognise this coexistence of contradictory versions, which “inhabit” and “coexist” in the very same space of knowledge.

  2. The rhizome, like the Encyclopaedia, is open: it has neither a definite beginning nor end, and each instance of the semiosis takes on new interpretations and new versions that the Encyclopaedia never stops registering. These new versions, these new interpretants constantly change the structure of the Encyclopaedia. And it is in this regard that Eco insists on the impossibility of a comprehensive and exhaustive description of the encyclopaedic network: both due to the fact that this description would have already changed it, and also due to the fact that an exhaustive description presupposes a point of view from the outside of the Encyclopaedia, which in his mind was completely impossible.

  3. The rhizome, like the Encyclopaedia, is constitutively non-cohesive, given that the mastery of the encyclopaedic contents changes according to sociolectal class, culture, individual etc..

    • 33 LLP: 44.

    The Encyclopaedia is an open labyrinth that cannot be observed from the outside, but can only be experienced from the inside. This is why Eco speaks of a “myopic algorithm”, since the image that one may have of the Encyclopaedia is only the myopic and partial image that is obtained by exploring the connections of knowledge from within. «There is no outside to a rhizomatic structure, therefore every perspective (every point of view on the rhizome) is always obtained starting from an internal point, hence every local description tends to be a mere hypothesis concerning the network as a whole. Inside a rhizome, thinking means proceeding by speculation».33

  • 34 SPM: 1029-1090.
  • 35 SFL: chapter 2.

29As it is an «open universe», which presents «a changeable form permeated with possibilities», always «in continuous expansion», Eco views the Encyclopaedia as a form of that idea of Adventure that was perfectly outlined in the Poetics of Joyce.34 Or rather, the Encyclopaedia has the form of holding together the Order of the Summa and the Adventure of the rhizome, because, if at a global level it defines a network in which every point can be connected to any other, at a local level it can assume ordered forms, which can be hierarchised and organised into intelligible forms of relationships.35 Further confirmation of a Thomism that has never stopped pulsing within even the most advanced models that Eco introduced: the local arborescent applications of the rhizomatic model of the encyclopaedia have the form of the summa. And this certainly constitutes a problematic aspect, which deserves further reflection.

6. From Encyclopaedia to Rhizome

  • 36 SFL: IX-XVII.

30Why do we have this strabismus between the general semiotic level, populated by rhizomatic models, and the local level of analysis, in which these very same rhizomatic models are actually ignored to the benefit of small local summae brimming in hierarchical and arborescent representations? According to Eco, semiotics is a discipline that has a dual soul:36 on the one hand, it is a philosophical and theoretical discipline that studies the nature of languages, on the other, it is an applicative discipline that uses theory tools in order to analyse languages that actually exist (texts). Hence the strabismus between theory and analysis. If, in fact, Eco described the contradictory nature of the semantic space perfectly at a general semiotic level, when he proceeded to apply his ideas to concrete cases, he made use of precisely those arborescent and hierarchical models that were rejected at a theoretical level. A paradigmatic example in this sense, for instance, is the analysis of metaphor in Semiotics and the philosophy of language.

  • 37 Personally, I tried to do so in my Strutturalismo e interpretazione.

31In short, if the encyclopaedia was a rhizome – non-hierarchical and centralised – this model was never actually used and Eco never actually built an applied interpretative semiotics that was consistent with his own theoretical assumptions.37 It was as if, in the tension between the Order of the Summa and the Adventure of the rhizome, his words lent towards the latter, but his actions continued undauntedly to pursue the former. However, there is an exceptionally profound theoretical reason that serves as a background of this decision, whereby the Theory can postulate the contradictory nature of the Encyclopaedia, but it must then render it Apollonian and urbanize it. The reason is provided in the Wittgensteinian paraphrase published on the cover of The Name of the Rose and then used as the central theoretical pivot of his philosophical autobiography (see LLP): «Whereof one cannot theorize, one must narrate» (NR). Eco explains it this manner in the Self-dictionary of Italian Authors.

  • 38 AD: 151.

Why in 1978 he decided to start writing a novel (The Name of the Rose, Bompiani, 1980), remains a matter, probably, for the psychoanalyst he never had. It is true that Eco, since the era of Verri, had ventured towards literary pastiches, later merged in Misreadings (Mondadori, 1963), but creating pastiches is not the same as writing a novel. Perhaps it had all been brewing for some time, and the author didn’t realise it. He didn’t realise it to the extent that, until it was published, he was convinced that he had created something totally foreign to his interests as a semiotician. It was actually his critics and readers who later pointed out to him the connections that existed between the two moments of his activity. Something that Eco could no longer ignore upon writing his second novel (Foucault’s Pendulum, Bompiani, 1988), which explicitly presents itself as the story of a neurosis of interpretation. In other words, he now believes that even writing novels is a moment of his philosophical activity, because there are questions that cannot be treated in an assertive essay manner, but must be staged, so to speak, in all their ambiguity and contradiction.38

  • 39 See Paolucci 2017b.

32This is a fundamental step. Well before the explicit admission in 2013 – with the launch of the Living Philosophers volume project that considers novels as a constitutive part of his philosophical activity – Eco already states very clearly to everyone that even his «novel writing» actually represents «a moment of his philosophical activity». The fundamental thing of this quotation is that it also tells us which part of his philosophical activity could not be handled by his Theory and had to be delegated to novels: the encyclopaedia’s contradictory nature should not be used inside the universe of Theory in order to construct models of analysis, but must be staged through stories in the universe of Narratives. The Self-dictionary is very clear: «There are issues that cannot be treated in an assertive essay manner», but «they must be staged in all their ambiguity and contradiction». Here is the profound meaning of «whereof one cannot theorize, one must narrate».39

33Eco is perfectly aware of the constitutive contradictoriness of what the semiotic theory had placed as its own object of analysis. If semiotics does not deal with the world, but with its versions, these versions of the world coexist in the semiotic space of the encyclopaedia and, from time to time, can be reactivated through the process of interpretation. There are two possible pathways to take from here: one can attempt to build non-hierarchical and non-centralised (rhizomatic) semiotic models, which can use the contradictory, open and non-cohesive structure of the encyclopaedia in the analysis. Or one can attempt to stage the «ambiguity» and «contradictoriness» of this very same space of coexistence of interpretations, in order to show its structure, at a different level. Eco choose the latter. The concrete activation of the encyclopaedic model are his novels: it is there that a rhizomatic structure is staged.

34It is as if Eco had felt somewhat embarrassed about thinking of a Theory constructed in an “unordered” form, a Theory that replicated the shape of that universe «in continuous expansion» which is the one of the Adventure. He postulates it, he says it, he is sincerely convinced that this is the case, but then, at the moment he renders the theoretical model of the Encyclopaedia operational, he does not personally take on all conceivable consequences and continues to construct orderly and arborescent representations. They are his small summae where, at a general semiotic level constructed in the image of the Adventure, a local application follows which leaves space for Order. In short, Eco attempts to distance himself from its Thomistic teachings throughout his entire intellectual adventure, but when he tries to describe the «reading rule», the «equation» that defines the Encyclopaedia as an «ever-expanding contradictory universe», he «finds in his hands» those «disguised formulas of the classical order» he had tried to distance itself from.

  • 40 See Bartezzaghi 2019; Ferraris, Paini 2018

35However, it is his novels that take charge of the Adventure concept, while Theory gradually abandons it, in order to concentrate on the Order concept. Theory in fact delegates Narration to carry out the construction and staging of the contradictory space of the encyclopaedia, which is not used to conduct scientific-academic analyses, but rather to build a series of stories that highlight the coexistence of contradictory versions of the world. Within these fictional worlds, in which the characters use encyclopaedia debris to construct lies and over-simplified explanations of a conspiracy-based style, Eco stages the forces of a destiny that marks his characters and accompanies them to their inevitable ruin (PF, B, CP, NZ). It is the irruption of that unamendable and necessary Truth, which is proper only to the narrative worlds (SL). It is as if, taking charge of the Adventure concept and the representation of the contradictions of the Encyclopaedia, Narration simultaneously took charge of that emancipatory function of philosophy to which Eco dedicated his semiological guerrilla (CC). As if to say: I know how it is easy to cheat, I know how to build representations that can be used to lie, I know how falsehood can produce a lot of true knowledge. But I know how to defend myself from this ambiguous universe, I know that the over-simplified explanations that circulate in mass culture do not contemplate the complexity of an expanding and hyperconnected universe, which is the one we are currently experiencing.40 And I am also well aware that conspiracy is the prototype of the oversimplified explanation, that which seeks a single cause for a multitude of events that do require many different explanations (NR). However, although I have tried to work on these topics at a semiological guerrilla and costume essay level, I unfortunately realise that very few even notice it and many are still victims of conspiracy-based fascination and easy explanations that hide the complexity of the contemporary world, which features thousands of interconnections that are not attributable to fake totalising explanations. Using stories for the general public, I can highlight the tragic destiny facing those who try to thrive on fake constructions (Belbo, Baudolino, Simone Simonini, Colonna) or those who live by the neurosis of interpretation (Diotallevi, Braggadocio). For this reason, for instance, in Eco novels, it is often only the characters showing common sense like Lia who emerge unscathed. And that is why this appeal to common sense will become increasingly important even in philosophical works, such as for example in Kant and the platypus (KO), where common sense plays a fundamental philosophical role, something which Eco himself explains to the reader from the outset.

7. A Conclusion in the Form of “Already Told”

  • 41 SFL: 261.
  • 42 Ibidem.

36In Semiotics and the philosophy of language, Eco41 recalls «that with the influence of the theory of Lévi-Strauss, the equation between social functioning and linguistic functioning, also referring to Jakobsonian linguistics», ended up imposing the reference to the notion of code all over the Humanities. Yet, this does not take place by metaphor or analogy, but rather in order to support the idea that even within mechanisms that are not immediately semio-linguistic and communicative, there is a mechanism of rules like those that have allowed us to understand language and communication. As Eco himself states,42 commenting on Lévi-Strauss’ idea of “kinship code”,

  • 43 SFL: 261.

the kinship-language equation does not really aim to demonstrate that interacting at a kinship level is communicating. Levi-Strauss is not saying that both kinship and language represent cases of communication; he suggests that society communicates also at the kinship level because there is a more general code which rules kinship, language, architectonic forms, and other phenomena. The point is: where there is rule and institution, there is society and a deconstructible mechanism. Culture, art, language, manufactured objects are phenomena of collective interactions governed by the same laws. Cultural life is not a spontaneous spiritual creation but, rather, is rule-governed. These rules represent an object of investigation. The concept of code serves not so much to suggest that everything is language and communication as to establish that every cultural production is rule-governed.43

  • 44 On this topic, see the discussion on Kantian Schematism summed up in Paolucci 2017c.

37The true objective of Eco is clearly evident, and represents the true essence of his philosophy and his thousands of other intellectual activities: when faced with an unknown phenomenon, one searches for the rule and an attempt to build a structural model that explains it is made, without any ontological hypostatisation of the model. This is why Eco states that even within non-semiolinguistic and non-communicative phenomena there is a mechanism of rules of the same type of those that allowed us to understand language and communication. But he never says that also non-linguistic and non-communicative systems have their own internal linguistic or communicative component or are structured as a language. Here is what Eco really cares for and was already at the core of the Open Work: being able to outline the conditions of passage and translatability between these different domains, each regulated by its own system of rules.44

38For this reason, the very same words that Eco used to describe the work of Lévi-Strauss can be applied to his own theory:

  • 45 SFL: 262.

Whilst fully aware of the adventurousness of his hypothesis, he warns us that it is by no means sufficient to limit the investigation to a single society, or even to many, if one does not identify a level where it is possible to move from one phenomenon to another. […] Therefore, during its first appearance, as already seen in Jakobsonian phonology, the code is presented not so much as a mechanism that allows communication but rather as a mechanism that allows the transformation between two systems. […] Linked to a communicative hypothesis, the code idea is not a guarantee of communication but of structural coherence, a carrier between different systems.45

39And this is exactly what Eco did with his historical analyses, his semiotics, his philosophy, his manner pieces and his novels. Exactly because research is an Adventure towards a new Order, our companions during this adventure will be the structures, the models, the codes and the rules which we use in order to fallibly attempt to make something intelligible, since we know we cannot explain it through intuition or through the Dionysian vortex of the creative spirit. This is the only way we can adequately respond to Peirce’s call to arms: i) we have no concept of the absolutely unknowable; ii) knowing something always means giving structured explanations. A call to arms to which Umberto Eco dedicated the totality of the Adventure of his thought.

Torna su


Barthes, R. 1970, S/Z, Paris, Seuil.

Bartezzaghi, S. 2019, Banalità, Milano, Bompiani.

Eco, U. AD: 1990, Autodizionario degli scrittori italiani, ed by F. Piemontese, Milano, Leonardo.

Eco, U. AL: 2007, Dall’albero al labirinto. Studi storici sul segno e l’interpretazione, Milano, Bompiani; From the Tree to the Labyrinth: Historical Studies on the Sign and Interpretation, tr. by A. Oldcorn, Cambridge (MA), Harvard University Press, 2014.

Eco, U. B: 2000, Baudolino. Milano, Bompiani; Baudolino, tr. by W. Weaver, New York, Harcourt, 2002).

Eco, U. CC: 1973, Il costume di casa. Evidenze e misteri dell’ideologia italiana, Milano, Bompiani.

Eco, U. CP: 2010, Il cimitero di Praga, Milano, Bompiani, Vintage ed., 2011; The Prague Cemetery, tr. by R. Dixon, Boston - New York, Houghton Mifflin Harcourt, 2011.

Eco, U. DM: 1963, Diario minimo, Milano, Mondadori, Rev. ed. Milano, Mondadori 1975 Milano, Bompiani; Misreadings, tr. by W. Weaver, New York, Harcourt Brace & Co., 1993; London, Cape, 1993.

Eco, U. FC: 1971, Le forme del contenuto, Milano, Bompiani.

Eco, U. FSS: 2015, La filosofia e le sue storie (with Riccardo Fedriga, eds), 3 vols, Roma-Bari, Laterza.

Eco, U. LLP: Intellectual Autobiography, in S.G. Beardsworth, R.E. Auxier, The Philosophy of Umberto Eco, vol. XXXV, Library of the living philosophers, Chicago, Open Court.

Eco, U. KO: 1997, Kant e l’ornitorinco, Milano, Bompiani; Kant and the Platypus: Essays on Language and Cognition, tr. by A. McEwen, London, Secker & Warburg, 1999.

Eco, U. NR: 1980, Il nome della rosa, Milano, Bompiani. Rev. ed. Milano, Bompiani, 2011; The Name of the Rose, transl. by William Weaver, San Diego and New York, Harcourt Brace Jovanovich, 1983.

Eco, U. NZ: 2014, Numero zero, Milano, Bompiani; Milano, Mondadori Retail, 2015; Numero Zero, tr. by R. Dixon, New York, Houghton Mifflin Harcourt, 2015.

Eco, U. OA: 1962, Opera aperta, Milano, Bompiani; 2nd, Rev. ed., 1967; English translations with other essays added: The Open Work, tr. by A. Cancogni, Cambridge (MA), Harvard University Press; London, Huchinson Radius, 1989.

Eco, U. PET: 1956, Il problema estetico in San Tommaso, Torino, Edizioni di Filosofia; 2nd, Rev. ed. Il problema estetico in Tommaso d’Aquino, Milano, Bompiani, 1970; The Aesthetics of Thomas Aquinas, tr. by H. Bredin, Cambridge (MA), Harvard University Press, 1988; London, Radius, 1988.

Eco, U. PF: 1988, Il pendolo di Foucault, Milano, Bompiani; Foucault’s Pendulum, tr. by W. Weaver, New York, Harcourt Brace Jovanovich, 1989.

Eco, U. PJ: 1966, Le poetiche di Joyce, Milano, Bompiani (Rev. ed. of the second part of Books 1962 Opera aperta): The Aesthetics of Chaosmos, tr. by E. Esrock, Tulsa, University of Tulsa, Monograph Series, 1982; Hutchinson, 1982. Republished as The Aesthetics of Chaosmos: The Middle Ages of James Joyce, London, Hutchinson, 1989; Cambridge (MA), Harvard University Press, 1989.

Eco, U. SA: 1968, La struttura assente. La ricerca semiotica e il metodo strutturale, Milano, Bompiani. Rev. ed., 1980.

Eco, U. SB: 2004, Storia della bellezza, Milano, Bompiani; History of Beauty, tr. by A. McEwen, New York, Rizzoli, 2004.

Eco, U. SBR: 2007, Storia della bruttezza, Milano, Bompiani; On Ugliness, tr. by A. McEwen, New York, Rizzoli International, 2007.

Eco, U. SFL: 1984, Semiotica e filosofia del linguaggio, Torino, Einaudi; Semiotics and the Philosophy of Language, Bloomington, Indiana University Press, 1984; London, Macmillan, 1984.

Eco, U. SL: 2002, Sulla letteratura, Milano, Bompiani; On Literature, tr. by M. McLaughlin, Orlando, Harcourt, 2004.

Eco, U. SPM: 2012, Scritti sul pensiero medievale, Milano, Bompiani.

Eco, U. SS: 1985, Sugli specchi e altri saggi. Il segno, la rappresentazione, l’illusione, l’immagine, Milano, Bompiani.

Eco, U. STL:2013, Storia delle terre e dei luoghi leggendari, Milano, Bompiani; The Book of Legendary Lands, London, MacLeose Press, 2013.

Eco, U. TSG: 1975, Trattato di semiotica generale, Milano, Bompiani; A Theory of Semiotics, tr. by Umberto Eco, Bloomington, Indiana University Press, 1977; London, Macmillan, 1977.

Eco, U. VL: 2009, Vertigine della lista. Milano, Bompiani; The Infinity of Lists, tr. by A. McEwen, New York, Rizzoli International, 2009.

Deleuze, G. 1969, De quoi on reconnaît le structuralisme, in F. Chatelet (ed.), Histoire de la philosophie, vol. VIII, Paris, Hachette; tr. it. part. in La filosofia del xx secolo, Milano, Rizzoli, 1973: 194-217.

Deleuze, G., Guattari, F. 1980, Mille plateaux. Capitalisme et schizophrénie, Paris, Les Éditions de Minuit.

Ferraris, M., Paini, G. 2018, Scienza nuova. Ontologia della trasformazione digitale, Torino, Rosenberg & Sellier.

Greimas, A.J. 1966, Sémantique structurale: recherche de méthode, Paris, Larousse.

Greimas, A.J. 1970, Du sens, Paris, Seuil.

Greimas, A.J. 1983, Du sens II. Essais sémiotiques, Paris, Seuil.

Greimas, A.J., Courtés, J. 1979, Sémiotique. Dictionnaire raisonné de la théorie du langage, vol. I, Paris, Hachette.

Jakobson, R. 1966, Saggi di linguistica generale, Milano, Feltrinelli.

Marmo, C. 2017, Eco’s Semiotics and Medieval Philosophy, in S. Beardsworth, R. Auxier (eds), The Philosophy of Umberto Eco, vol. XXXV, Library of the living philosophers, Chicago, Open Court: 103-131.

Marrone, G. 2010, L’invenzione del testo, Roma-Bari, Laterza; The Invention of the Text, London, Mimesis International 2014.

Paolucci, C. 2010, Strutturalismo e interpretazione, Milano, Bompiani.

Paolucci, C. 2016, Whereof One Cannot Theorize, Thereof One Must Narrate, in T. Thellefsen, B. Sørensen (eds), Umberto Eco in His Own Words, Berlin, Boston, De Gruyter: 165-173.

Paolucci, C. 2017a, Umberto Eco. Tra Ordine e Avventura, Milano, Feltrinelli.

Paolucci, C. 2017b, Eco, Peirce and the Anxiety of Influence. The Most Kantian of Thinkers, in S. Beardsworth, R. Auxier (eds), The Philosophy of Umberto Eco, vol. XXXV, Library of the living philosophers, Chicago, Open Court: 103-131.

Paolucci, C. 2017c, Semiotic, Schemata, Diagrams, and Graphs: A New Form of Diagrammatic Kantism by Peirce, in K. Hull, K. Atkins, (eds), Peirce on Perception and Reasoning, Routledge: 74-85.

Paolucci, C. 2021, Cognitive Semiotics: Integrating Signs, Mind, Meaning and Cognition, Dordrecht, Springer.

Rosenstiehl, P., Petitot, J. 1974, Automate asociale et systèmes acentrés, “Communications”, 22: 45-62.

Violi, P. 1997, Significato ed esperienza, Milano, Bompiani; Meaning and Experience, Bloomington, Indiana University Press, 2001.

Torna su


1 The Aesthetics of Thomas Aquinas was originally written in 1956. The foreword was written in 1988.

2 As it is well known, Peirce laid the foundations of his pragmaticism on a semiotic theory which gave orientation to human actions. See Paolucci 2021.

3 SFL: 300-301.

4 NR: 345.

5 See Paolucci 2017a, chapter 2.

6 PJ: 1044.

7 In The Role of the Reader, in a translation of the paragraph on the “semantics of metaphor” originally appeared in Le forme del contenuto (1971) Eco explicitly refers to its linguistic, lexical and, thus, semantic nature.

8 PJ: 1043-1044.

9 PJ: 1043.

10 See

11 For a more detailed and broader overview of the connections between Eco’s semiotics and mediaeval philosophy, see Marmo 2017.

12 SPM: 1245.

13 Ibidem.

14 On this point, see the famous chapter of Jakobson’s Writings in general linguistics.

15 On these topics, see the extended discussion in Strutturalismo e interpretazione (Paolucci 2010), chapters 1 and 3.

16 See Violi 1997.

17 See Greimas, Courtés 1979, “Text”, see also Marrone 2010.

18 See Deleuze 1969.

19 SFL: 265.

20 SF: 282.

21 TSG: 13.

22 SFL: 281.

23 SFL: chapter 5.

24 TSG: 56, 74, 77, 179.

25 See Paolucci 2010, chapter 4.

26 SFL, chapter 2.

27 Barthes 1970: 24.

28 SFL: 295.

29 See Paolucci 2016, 2017a.

30 SS: 359-360.

31 SS: 334-9.

32 LLP: 44.

33 LLP: 44.

34 SPM: 1029-1090.

35 SFL: chapter 2.


37 Personally, I tried to do so in my Strutturalismo e interpretazione.

38 AD: 151.

39 See Paolucci 2017b.

40 See Bartezzaghi 2019; Ferraris, Paini 2018

41 SFL: 261.

42 Ibidem.

43 SFL: 261.

44 On this topic, see the discussion on Kantian Schematism summed up in Paolucci 2017c.

45 SFL: 262.

Torna su

Per citare questo articolo

Notizia bibliografica

Claudio Paolucci, «The Notion of System in the Work of Umberto Eco: Summa, Structure, Code, Encyclopaedia and Rhizome»Rivista di estetica, 76 | 2021, 39-60.

Notizia bibliografica digitale

Claudio Paolucci, «The Notion of System in the Work of Umberto Eco: Summa, Structure, Code, Encyclopaedia and Rhizome»Rivista di estetica [Online], 76 | 2021, online dal 01 mai 2021, consultato il 18 juin 2021. URL:; DOI:

Torna su

Diritti d’autore

Licenza Creative Commons
Rivista di Estetica è distribuita con Licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 4.0 Internazionale.

Torna su
  • Logo Rosenberg & Sellier
  • DOAJ - Directory of Open Access Journals
  • OpenEdition Journals
Cerca su OpenEdition Search

Sarai reindirizzato su OpenEdition Search