Semiotics for Actor-Network Theory

Abstract

This chapter aims to show how the relationship between Latour's investigation on meaning and the research perspective on signification developed within the framework of structural semiotics is anything but episodic, marginal or outdated. With the aim of highlighting the reasons behind an uninterrupted dialogue, even if often denied or marginalised in the field of social sciences, this chapter considers the main semiotic concepts at the basis of Latour’s work, highlighting the main affinities and discontinuities that emerge at a theoretical and methodological level, with particular reference to actor-network theory. It is pointed out that the metaphor of semiotics as a toolbox for actor-network theory is overly reductive and does not sufficiently account for the presence of a common non-anthropomorphic theory of agency and an anti-dualist epistemological principle that recognizes the primacy of the relationship over the elements involved in a social phenomenon. To this end, a comparison is made between the notions of actor, actant, enunciation, narrative program originally elaborated within the framework of semiotic theory and their novel reinterpretation introduced by Latour to examine the paradoxes of modernity.

3.1 Much More than a “Toolbox”

Despite the relationship between the investigation on meaning advanced by Latour and the research perspective on signification developed within the framework of structural semiotics being far from episodic, weak or marginal, the overall contribution of the latter has been scarcely recognized¹ within the social sciences. As Fabbri himself (my trans.; 2021: 34) clarifies, this is proof of the reasonable effectiveness of semiotics and at the same time, its presumed “outdatedness”:

A sort of outdatedness of semiotics, notably the diffusion of Greimas’ theoretical proposal is also due to its success, to the fact that many of his concepts have overflowed, that they are around, are widely used […] without their genealogy being cited, without recognizing the paternity of those who put them into circulation.

The situation is different within semiotic studies, where in recent years there have been multiple signs of renewed interest for Latour’s work and the perspectives arising from it in relation to the investigation of the different ways of existence and the phenomena of signification at play in society (Padoan, 2023; Peverini, 2021, 2023a, b; Mangano, 2021; Ventura Bordenca, 2021; Mattozzi, 2021; Finocchi et al., 2020; Paolucci, 2010, 2020; Sedda, 2021; Burgio 2021; Lorusso, 2020; Manchia, 2020; Marrone, 2011, 2019; Fontanille & Couégnas, 2018; Fontanille, 2014; D'Armenio, 2019; Dondero, 2017; Tassinari, 2017).The attention paid in contemporary semiotic research to Latour’s work is expressed on two dimensions that are interrelated. The first concerns the need to overcome the prejudice that reduces the contribution of semiotics to the social sciences to that of a “useful toolbox” for actor-network theory. The second concerns the need to recognise how, over time, Bruno Latour’s original work of theoretical expansion around concepts such as actant and enunciation has proven fruitful in advancing semiotic reflection on the way meaning is articulated and manifested in a contemporaneity increasingly marked by a multiplicity of relations between human and non-human actors, proving to be a “fruitful source of epistemological, theoretical, methodological and empirical inspiration” (my trans.; Marrone, 2010: 72).

In order to highlight the reasons for an uninterrupted dialogue, although often denied or marginalized within the field of social sciences, in this chapter the main semiotic concepts underlying Latour’s work are considered, with the aim of highlighting some of the main affinities and discontinuities that emerge on the theoretical and methodological level, with particular reference to the actor-network theory.

Starting from Laboratory Life, a text published in 1979 together with sociologist Steven Woolgar, and continuing with the publication of Les Microbes: guerre et paix, suivi de irréductions (1984), Science in Action (1987) and Nous n’avons jamais été modernes (1991), Latour strengthens the conviction that it is necessary to overcome the asymmetry that assigns a priority role to the social explanation of phenomena, underestimating the role assumed by non-human agents. His proposal therefore consists in introducing a principle of general symmetry according to which the investigation of natural objects and that of the social must proceed simultaneously.

Neither nature nor society can thus be considered as the foundation or the guarantee of the stabilization of scientific knowledge. Rather, “purely” natural facts and “purely” social facts would be the effect of a subsequent process of dissimulation, called purification, capable of hiding the complex intertwining of relationships that makes possible the birth of an object (whether theoretical, technical or natural), distinguishing only afterwards what belongs to the order of the natural and what refers to the dimension of the social.

The principle of symmetry is not only generalized but also radicalized and it is here that the actor-network theory takes shape,² starting from the conviction that all actors, who in the set of links they establish among themselves contribute to the constitution of a scientific phenomenon (agency), must be considered on the same level regardless of whether or not they are human.

What characterizes this approach is therefore not the attempt to explain scientific facts and technical innovations in light of social dynamics. The goal rather consists in reconstructing, through procedures of observation and meticulous description, the way in which concepts, natural objects, and technical objects emerge, take shape, and stabilize over time, now understood as network-actors, that is, as collective actors composed of a multitude of heterogeneous components whose functioning is based on a network-like structure.

To account for the construction of the phenomena investigated, it becomes essential to reconstruct the circulation of all the elements endowed with agency—the actants—that enter into relation among them, giving rise to a series of transformations, to a network of translations that makes the manifestation of an actor possible. A paradigmatic example of this constructivist approach is represented by the work dedicated by John Law (1984) to the birth and evolution of the Portuguese commercial empire in the 1500s. The premise from which this study starts is that there are two perspectives for analysing such phenomenon. On one hand, the consolidated historical approach, which provides an explanation of the object of investigation in the form of a conventional narrative focused on recurring key actors: spices, trade, wealth, military power, and Christianity, relegating the technological factor, although considered essential, to a marginal role. On the other hand, the perspective typical of maritime and naval history which, conversely, examines the innovations in shipbuilding and navigation, paying less interest to the political and economic dynamics at play in Portuguese imperialism.

Law’s proposal consists in combining these two types of historical narrative starting from the consideration that the effectiveness of the Portuguese commercial empire is attributed to the assembly of multiple actors: technical (sails, astrolabes, firearms etc.), natural (winds, currents, constellations, spices etc.), commercial (shops, commercial agreements etc.), circulating within a system of network-like translations. And it would have been precisely the network generated by the transformations of the individual elements that determined the particular form of each actant involved in its operation, ensuring its hold for 150 years and securing for Lisbon a hegemonic role within an asymmetric commercial structure that made the city a mandatory stopover for a vast array of tributaries.

This example is useful for grasping the originality of ANT on a theoretical-epistemological level and for overcoming some misunderstandings that are often associated with it. As Mattozzi (2006) points out, this research perspective is indeed subject to two frequent misunderstandings, the first of which consists in reducing it to the theory of social constructivism, the second in considering the terms actor-network as dichotomous and therefore interchangeable with other variants such as individual/system or agency /structure:

ANT cannot be ascribed to social constructivism, since it does not consider society as something that is given a priori. It is a constructivism, but not social [...]. Actor-network is a compound word that accounts for the fact that an actor is always the result of a network of relationships that constitute it. (my trans.; Mattozzi, 2006: 45)

The hyphen that separates the word actor from network thus responds to the need to indicate that the two terms do not express separate concepts but rather refer to two different aspects of the same field of observation and analysis. Speaking of an “actor” means focusing the research work on how networks are constituted by the action of the nodes that compose them; conversely, the use of the term “network” is useful to describe the ways in which individual actors are defined by the ties that are established with the other elements involved in the functioning of a phenomenon. From this perspective, the distinction mainly responds to a methodological need but does not introduce any ontological distinction between the two terms.

Moreover, the relevance assigned to the methodological dimension is a central aspect in the reflection of the scholars who have mainly contributed to the birth and development of ANT and who significantly agree in recognizing that, despite its use, the term "theory" is very misleading,³ as provocatively claimed by Latour himself (1999b: 1; 19–20)

[…] there are four things that do not work with actor-network theory; the word actor. the word network, the word theory and the hyphen! Four nails in the coffin […] The third nail in the coffin is the word theory […] Far from being a theory of the social or even worse an explanation of what makes society exert pressure on actors, it always was, and this from its very inception (Callon & Latour, 1981), a very crude method to learn from the actors without imposing on them an a priori definition of their world-building capacities.

ANT should therefore be considered, rather than as a social theory, as a methodology, or more precisely as a meta-methodology (Venturini, 2008) that uses a multitude of different research approaches (interviews, text analysis, archival work, surveys, experiments, participant observation) to try to describe the work of association and dissociation that permeates collective life.

In this scenario, the innovative scope of ANT is also due to the presence of a dialectic with semiotics, a comparison made of affinities and distinctions but always played against the backdrop of a common anti-dualist epistemological horizon which consists in recognizing the primacy of the relationship with respect to the elements involved in the unfolding of a phenomenon.

The following paragraphs are dedicated to exploring the areas of contiguity and divergence between two ways of reflecting on the social that, while distinguishing themselves by the different importance assigned to methodological elaboration compared to empirical verification and to the descriptive effort towards the phenomena investigated, share a series of basic principles.

In particular, an attempt will be made to highlight how the metaphor of semiotics as a toolbox is very reductive and does not sufficiently account for the fundamental role played by the theory of signification in the elaboration of ANT as very well argued by Høstaker (2005) and Beetz (2013). In other words, an attempt will be made to clarify how the significant impact generated by ANT also testifies to the effectiveness of semiotics understood as a “methodology for social sciences” (my trans.; Fabbri, 2023). Following this perspective, an attempt will be made to show how the originality of Latour’s work can be traced back, at least in part, to an unprecedented work of bricolage⁴ carried out starting from Greimas’ theoretical framework, with the aim of extending the relational epistemology to non-linguistic phenomena.

One aspect that indeed emerges as particularly worthy of interest is that ANT’s recourse to semiotics, while materializing in the use of a relatively small number of terms, produces the effect of advancing reflection on the scope of the selected conceptual categories, expanding their scope on a theoretical level.

3.2 A Non-anthropomorphic Theory of Agency

The relevance of the bricolage metaphor clearly emerges in the fine-tuning of a semiotic vocabulary (Akrich & Latour, 1992) used to account for the concatenations of humans and non-humans, in which the recourse to some foundational concepts of Greimas’ semiotics is accompanied by a redefinition of them, as in the case of the definition of the key category of actant.

Whatever acts or shifts action, action itself being defined by a list of performances through trials; from these performances are deduced a set of competences with which the actant is endowed; the fusion point of a metal is a trial through which the strength of an alloy is defined; the bankruptcy of a company is a trial through which the faithfulness of an ally may be defined; an actor is an actant endowed with a character (usually anthropomorphic) (Akrich & Latour, 1992: 259).

This definition allows us to understand how this “theory” of action is based on a basic assumption that consists in rejecting intentionality as a useful criterion for identifying and distinguishing the heterogeneous entities involved in the unfolding of a phenomenon, in favour of the centrality assigned to their ability to act (agency). What qualifies an entity as an actant is therefore not the will to do but the ability to perform an action capable of making a difference, and in this sense the term can be used without distinction with reference to human beings, technological artifacts, natural elements, institutions, legal norms, concepts, etc.

In this perspective, the notion of actant thus assumes a decisive role, because it can account for the way in which, in the structuring of a social phenomenon, agency is redistributed among a multiplicity of interconnected elements, manifesting itself in the form of a chain. What distinguishes an actor from an actant is therefore not its anthropomorphic dimension but rather the ability to consolidate the presence of a plurality of multiform entities, to concretize an assemblage (Croce, 2020).

The relevance assigned to the notions of actor and actant in Latourian thought is also proven by the different definitions of the terms elaborated over time and that testify to an action of “creative remodeling” carried out starting from Greimas’ semiotic theory (Beetz, 2013).

As Beetz points out (2013: 9) if in some works like Where are the Missing Masses (1992) and Politiques de la nature (1999a) the definition of actor is very similar to the one just presented, in others the distinction between actor and actant is decidedly blurred, as in the case of Science in action (1987), or completely absent as in the 1984 essay Les microbes; in Reassembling the social (2005) what qualifies an actor is the ability to make a difference, while the actant is defined as an actor who is still without a concrete configuration. A different perspective of reflection is finally found in Pandora's Hope (1999c) in which the term actant is used only occasionally to refer to non-humans and what becomes relevant is the reflection on the way in which an actor progressively emerges as a result of the ability to overcome a series of trials.

In any case, beyond a certain flexibility in the use of the categories of actor and actant, the link with semiotics is very evident, especially with regard to the overcoming of an anthropomorphic vision of agency and, as anticipated, the common adherence to a relational and differential principle at the basis of signification. Regarding the first point, in the definition by Greimas and Courtés (1982: 5), the actant is indeed conceived as “that which accomplishes or undergoes an act, independently of all other determinations”, the term is introduced in fact to overcome another one considered excessively ambiguous: that of character. The category of actant is thus used to designate everything that participates in the functioning of a narrative and that can assume a concrete manifestation (actor) through a multiplicity of forms: human being, animal, object but also concept.

As for the second point, the perspective advanced by Latour and Akrich according to which in the functioning of a phenomenon actors and actants do not exist and never act in isolation but always in combination with each other, while being consistent with the approach of structuralist semiotics, is characterized at the same time by a certain theoretical simplification. In the ANT project, the extensive use of the notion of actant is functional to the project of describing the way in which, outside the linguistic dimension, a phenomenon of collective interest is based on a chain of transformations that involve a multiplicity of heterogeneous elements endowed with the capacity for action. According to this perspective, as emerged from the first works dedicated to the rhetoric of the scientific text, the strength of the theoretical proposal is measured primarily on the ability to trace the actants involved in the process of redistributing agency.

In the semiotic approach, the term actant is instead part of a complex and broader theoretical elaboration (the narrative grammar) whose objective is to reconstruct the formal logics of the functioning of narrativity. In particular, the word actant designates an abstract category and refers to a conception of narrativity understood as syntax, in which what is relevant is not the identification of the actants (as happens in ANT) but primarily the position they occupy within the transformations in which they are involved, and the (actantial) roles they exercise within a text, as clarified by Greimas and Courtés (1982: 6):

As the narrative discourse progresses, the actant may assume a certain number of actantial roles, defined both by the position of the actant in the logical sequence of the narration (its syntactic definition) and by its modal investment (its morphological definition). Thus the hero will be the hero only in certain parts of the narrative—s/he was not the hero before and s/he may well not be the hero afterwards.

In other words, the relevance of the notion of actant in the perspective of semiotic theory consists in the fact that it is a syntactic unit whose functioning responds to a generative type of signification logic. That is to say, a model in which the components that intervene in the production of an object endowed with meaning articulate with each other according to a path that goes from an elementary level to a more complex one, from an abstract plane to a concrete one, consistently with the conception of language levels (Benveniste) and of semiotics considered as a hierarchy (Hjelmslev). Furthermore, the narrative grammar (of which the actants are one of the key components) is a general model of narrativity, developed to account for the functioning of all forms of narrative text, regardless of the natural language in which they manifest. In this sense, the actantial organization is applicable to any form of narration and semiotic analysis is characterized on the methodological level precisely by the option of segmenting and analysing the text in terms of its structure, using therefore as a fundamental criterion the respect for the levels on which the articulation of the narrative is based. Particularly significant in this regard is the botanical metaphor used by Greimas and Courtés (1982: 258) to account for the procedures of reduction that necessarily characterize semiotic analysis: “here the semiotician can be compared to the botanist, whom no one would criticize for bracketing out in his work the aesthetic or economic aspects of the flowers which he studies”.

Here emerges a significant difference between Greimas’ theory, often superficially accused of reductionism,⁵ and the approach of ANT, which resolutely refuses to reduce the chain of transformations that occur between a multitude of actants to a series of general operating principles organized according to a hierarchical logic. This is what Madeleine Akrich programmatically states in an essay titled Sémiotique et sociologie des techniques: jusq’où pousser le parallèle? (my trans.; 1992):

Compared to a simple network model, semiotics poses an additional hypothesis by establishing a priori a characterization and a hierarchy of different elements put in relation: it is clear that it is not possible to accept these assumptions that go against our methodological hypothesis of departure, according to which the only way to reconstruct the network of relationships woven by a technical object consists in following the actors in their work of concatenations and not to impose any category, nor any link, that is not made effective by one of the actors in situation.

Certainly, this difference refers, as anticipated in the first chapter, to the constitutive project of STS and ANT to account for the signification at play in phenomena, such as technical objects, deeply distinct from traditional textual objects such as narratives, from which, as we know, the semiotics of narrativity has developed its own models of analysis. However, it is important to highlight how this distancing reveals a discord that is situated on the methodological level rather than the epistemological one, as is quite evident from the previous quote by Madelein Akrich herself and in a statement made by Latour in Reassembling the social.⁶

In particular, the refusal to resort to the general model of structuralist-inspired narrativity does not at all imply the abandonment of the anti-essentialist principle, central in semiotics, rather it responds to the attempt to reconcile the theory of signification with the American ethnomethodology characterized by a phenomenological perspective, in an attempt to account for the processes of modernization. As we will see extensively later, the move advanced by Latour in the famous essay We Have Never Been Modern indeed involves extending the analysis of the construction of scientific facts to the phenomena of modernity, with the aim of demonstrating that our everyday life is populated by hybrid actors, networks composed of human and non-human agents that only naturalistic or sociological reductionism persists in separating and considering as belonging to irreconcilable poles, resorting to dichotomies such as natural/artificial, subject/object.

[…] when we find ourselves invaded by frozen embryos, expert systems, digital machines, sensor equipped robots, hybrid corn, data banks, psychotropic drugs, whales outfitted with radar sounding devices, gene synthesizers, audience analyzers, and so on, when our daily newspapers display all these monsters on page after page, and when none of these chimera can be properly on the object side or on the subject side, or even in between, something has to be done (1993b: 49–50).

From this perspective, the scenario of everyday life is permeated by the incessant production of nature-cultures that Latour calls collectives, a term introduced to assert that these are phenomena that do not coincide either with the conception of society elaborated by sociology (“humans among themselves”) nor with the epistemologists’ idea of nature (“things in themselves”).

What ANT claims is the opportunity to carve out a space for observation and description of the phenomena of modernity that remains at the same distance between realism and constructivism and that therefore allows us to observe the procedures through which nature and society constitute each other.⁷ If the reality in which we live is the product of a process of construction and temporary stabilization that is realized through immanent relations between a multitude of agents (human and non-human) giving life to networks composed of hybrids, the challenge consists in reconstructing the interweaving of relations between the entities involved by describing the modes of association and of translation.

The dialectic with semiotic theory here proves to be as heated and stimulating as ever. On the one hand, in fact, the theory of signification is recognized as having the ability to offer an “excellent tool chest for following the mediations of language” (1993b: 64) to reconstruct the network of translations that makes possible the proliferation of hybrid agents, enabling an escape “from the parallel traps of naturalization and sociologization” (ibid). On the other hand, Latour asserts that the innovative scope of first the linguistic turn, then the semiotic one, would have been halted due to the choice to excessively expand the space assigned to the autonomization of discourse, in relation to the pole of nature and that of the subject/society, ending up progressively marginalizing both the question of the referent and that of the link with the speaker and the social context. In short:

[…] the great weakness of these philosophies, however, is to render more difficult the connections between an autonomized discourse and what they had provisionally shelved: the referent—on Nature's side—and the speaker—on the side of society/subject. Once again, science studies played their disturbing role. When they applied semiotics to scientific discourse, and not only to literatures of fiction, the autonomization of discourse appeared as an artifice (Bastide, in press). As for rhetoric, it changed its meaning entirely when it had truth and proof to absorb instead of conviction and seduction (Latour, 1987). When we are dealing with science and technology it is hard to imagine for long that we are a text that is writing itself, a discourse that is speaking all by itself, a play of signifiers without signifieds (ibid).

This position, although openly critical, should not be read in any case as a sort of “condemnation”, also because it intercepts a debate that has strongly marked semiotics since the end of the eighties, the positive outcome of which, as we will see, was to “rethink the real as the other side of the textual” (my trans.; Landowski, 1989), reinterpreting textuality, through the elaboration of sociosemiotic theory, “not as an objective entity given to imitate the literary work, but as a battery of formal models, structural grid, plot through which meaning is put in conditions to signify” (my trans.; Marrone, 2010: 72).

As can be seen from these passages, the metaphor of the toolbox does not do justice to the dialectic between semiotics and ANT which, particularly with Latour, is never reduced to the use by the latter of a number (certainly reduced) of categories but consists in an operation of deepening and expanding the theoretical value of the concepts used. What is particularly important to ANT is to practice semiotics understood as the study of how meaning is built “in its original, nontextual and nonlinguistic interpretation […]”, more precisely as “the study of order building or path building and may be applied to settings, machines, bodies, and programming languages as well as texts” (Akrich & Latour, 1992: 259).

Returning to the question of actants and actors, their use allows Latour to equip the process of observing social phenomena with a metalanguage that inevitably relies on the theory of narrativity (of which these two concepts are an essential component). Just as the theory of narrativity elaborated by structuralist semiotics implies at its foundation a theory of agency, according to which narrative roles are defined by the positions occupied by the characters in a chain of transformations, the solution adopted by ANT in the study of the way scientists progressively construct their research objects consists in describing the forms or the types of actions the actants are involved in, regardless of the level of manifestation (anthropomorphic or not) with which they manifest themselves. The link between the two approaches is further evident in the moment when the common recourse to a theory of agency centred on the polemical-contractual relations between the actants emerges.

Starting from Propp’s (1928) studies on the morphology of the folktale, Greimas’ semiotics (Greimas & Courtés, 1982) indeed conceives narrativity as a path organized in a canonical schema composed of four phases in which the narrative roles exercised by actants respond to a model based on a general logic that assigns a decisive role to conflict and alliances (Figs. 3.1 and 3.2).

Fig. 3.1

Canonical narrative schema

Fig. 3.2

Actantial model

Put simply, in the canonical narrative schema the manipulation phase is defined as the one in which a Sender convinces a Receiver to act in a certain way or to desist from a behaviour, adopting persuasive strategies such as promise, threat, seduction or provocation. In the second phase, the competence, the Subject (the hero of the narrative), is called upon to acquire a series of skills necessary to complete the task, to join with the object of value. The abilities that the Subject must develop not only concern the pragmatic dimension of doing but also the pathemic sphere, that is the emotions at stake, and the cognitive one, that is the knowledge gained during the journey. To carry out the narrative program the Subject must come into possession of some modalities. In semiotic theory, modalities are identified in function of four essential modal verbs: wanting (vouloir), having-to do or to be (devoir), being-able to do or to be (pouvoir) and knowing (savoir) which refer to the basic orientations that regulate the relationship between the subjects and between the subject and the world. The “having-to” concerns the sphere of social coercions, the “wanting” refers to the realm of desires that move the subject, the “knowing” refers to the knowledge necessary to undertake an action, finally the “being-able to” is relative to the concrete possibility to carry out an action.

The subsequent phase of the performance consists in the attempt by the Subject to carry out the program that defines his/her role within the narrative. Finally, in the concluding phase of the sanction, the path taken by the Subject is judged by the Sender, an evaluation of the path taken that can be positive or negative.

As is evident, from the structuralist perspective the narrative is thus conceived as a narrative syntax, as a sequence of states and transformations in which the status of the involved elements is defined by the set of trials in which they are implicated.

And it is precisely the role assumed by the trials in the functioning of a narrative that assumes a relevant role in the reflection of Latour, who in an innovative way proposes to extend the range of a central category in semiotics by changing its field of application, moving from the literary field to that of experimental sciences.

3.3 Anti-reductionism and Relational Ontology

The premise from which Latour begins is that if by actant we mean everything that has the ability to act by causing a change and if actions are defined as a “list of performances in trials”, in the absence of trials it is not possible to identify actors.

Now, this is exactly what happens in a laboratory where the protocol that scientists are required to follow consists in subjecting an “object” to a series of trials for the purpose of understanding its operation. According to this perspective, it is the so-called “trials of strength” at play in scientific practice that determine the endurance of a technical or scientific object, defining its status, determining whether or not to qualify it as a full-fledged actor, thus substantially defining its ontological status. This is the principle of irreduction according to which “it is because nothing is, by itself, reducible or irreducible to anything else that there are only trails (of strength, of weakness). What is neither reducible nor irreducible has to be tested, counted, and measured. There is no other way” (Latour, 1988: 158).

In this constructivist vision of the scientific fact, the entity that has yet to be subjected to a series of trials (in the form of experiments) initially does not possess an identity, is therefore not an actor, in the sense that, as Høstaker (2005: 10) reminds us, it can only be described as “the name of action” (Latour, 1993a: 136, 1999c: 119), as a “list for a series of trials” (Latour, 1987: 89).

The entity acquires an identity, becoming a stabilized and identifiable scientific object, only when the performances manifested during the set of tests to which it is subjected are recognized by scientists as the prerequiste of a competence that retrospectively explains its functioning. This also implies that the identity of a scientific object, its legitimacy, can be downsized or vanish altogether when it fails to confirm its own performance in a series of tests following its identification.

The emblematic example proposed by Latour (1987) is that of the discovery, in the early twentieth century, of “N rays” by the French physicist Prosper-René Blondlot. This led to numerous scientific publications in authoritative journals, and practical applications even in the field of medicine, until the moment when the American scientist Robert W. Wood, who had failed to replicate Blondlot’s experiment, decided to visit his laboratory. After an initial phase in which the existence of “N rays” seemed empirically unassailable, Wood, who in the meantime had asked to access the technical equipment used for the discovery, decided to surreptitiously remove the crucial element of the experiment: an aluminium prism essential for measuring the performance of the rays. Surprisingly, he observed that this did not imply any variation in the results obtained (traces imprinted on metal plates). The inscriptions were therefore not caused by “N rays” but by something else. Latour’s conclusion is that after the failure of this test, no one saw the “N rays” imprinted on the photographic plates anymore, only smudges.

The application of notions derived from a general theory of narrativity to the logic of science in action is relevant not only because it allows the semiotic dimension of scientific practices to emerge, but also to grasp a peculiarity, namely the inversion of the logical relationship that exists between the phases of competence and performance. In fact, while in the analysis of a narrative competence precedes performance, grouping all the intermediate (qualifying) stages that a subject goes through in order to realise its main narrative programme, scientific practice is characterised by the opposite movement. The observation of performance always precedes the reconstruction of competence, and, as intuited from the first works dedicated to the rhetoric of scientific discourse, the argumentative effectiveness of the texts in charge of attesting the reliability of the results obtained in the laboratory is played precisely on the ability to convincingly recount, a posteriori, the concatenation of states and transformations through which a scientific object progressively assumes a form, retracing the history of the experiment, trying to represent it in a way that hinders potential disputes from opponents.

Furthermore, from the perspective of ANT, the relevance assigned to trials of strength far exceeds the scope of technical or scientific objects. This concept in fact represents a fundamental principle through which to describe everyday life, the modes of existence of collective actors, concepts, theories, political institutions: “there are only trials of strength, of weakness. Or more simply, there are only trials. This is my point of departure: a verb, “to try” (Latour, 1988: 158). If the reality of an actant is sanctioned by the ability to resist a series of tests, a decisive aspect consists in its ability to enter into a relationship with other actants, giving rise to associations and alliances. It is in this passage that the idea of meaning as a trajectory takes shape.

No “actant” is ever so weak that it cannot enlist another. Then the two join together and become one for a third actant, which they can therefore move more easily. An eddy is formed, and it grows by becoming many others” (Latour, 1988: 159).

The centrality of the relational principle re-emerges here, once again making explicit the link between ANT and the theory of signification on the epistemological level. Just as the capacity of a language to express meaning depends on the system of relations established between its elements, so material entities too acquire meaning solely as a function of the links established with other entities.

At the same time, however, it is important to reiterate that in this perspective what is lost is the primacy of the notion of system and with it the assumption of a hierarchy of elements that presupposes rigid rules of concatenation.

[…] the notion of system is of no use to us, for a system is the end product of tinkering and not its point of departure (2.1.4). For a system to exist, entities must be clearly defined, whereas in practice this is never the case; functions must be clear, whereas most actors are uncertain whether they want to command or obey; the exchange of equivalents between entities or subsystems must be agreed, whereas everywhere there are disputes about the rate and direction of exchange. Systems do not exist, but systematizing is common enough; everywhere there are forces that oblige others to play the way they have always played (Latour, 1988: 198).

In the perspective advanced by Latour, the association between actants always responds to a polemical logic aimed at ensuring the strongest the ability to persist over time and extend in space, in this sense the outcome of an assemblage of distinct units cannot be predicted with certainty. There is therefore an inverse relationship between the extension of networks and the coherence of actors, since the fate of a trajectory of meaning is always played between two opposing forces: on one hand risking “dissidence” or “dissolution”, that is, jeopardizing one’s own uniformity to try to extend far, on the other hand strengthening one’s own coherence, resisting the other forces in play but at the same time reducing the possibilities of expanding, thus jeopardizing one’s own hold over time.

The solution from a methodological point of view consists in trying to trace the movements that regulate the relations between the actants involved in a social phenomenon, in search, as Croce (my trans.; 2020: 26–27) states, of “‘unique’ explanations, that is—as the reductionist principle commands—such that they cannot be abstracted from the single event and applied to others”.

3.4 From Narrative Programs to Action Programs

For ANT, the ability of an actor to manifest a goal or, in the absence of intentionality, to nevertheless operate a transformation in relation to a pre-existing situation prefigures the existence of a program of action. This term designates the set of trials and polemical comparisons that an actor⁸ faces to carry out his/her own program, measuring himself/herself against the resistance expressed by antagonists (anti-actors).

Significantly, the elaboration of this category also explicitly refers to the theoretical framework of semiotics, particularly to the notion of narrative program (NP) that Greimas and Courtés (1982: 245) define as “an elementary syntagm of the surface narrative syntax, composed of an utterance of doing governing an utterance of state […] as a change of state effected by any subject (S1) affecting any subject (S2)”.

This notion plays a key role in the functioning of the overall architecture of a narrative and can take on different configurations. In particular, a base NP (the goal that an actor aims for) is distinguished from an instrumental NP (e.g. the intermediate actions necessary to equip the subject of the narrative with the being able to do necessary to fulfill his/her role). The latter can be realized either by the subject himself or by another subject delegated by the first, giving rise to an annex NP.

In the lexicon developed by Latour, rather than usage NPs and annexed NPs we find deviations and delegations which nonetheless refer to a similar concept of agency. Also in this perspective, the ability of actors to contribute to generating a transformation often turns out to be anything but linear, being marked by a sequence of changes of direction necessary to complete intermediate programs and by the transfer of programs between the various components involved.

A distinctive feature introduced by ANT compared to semiotic theory instead concerns the central role assigned to the associations that are established between different actors and, in particular, to the formation of hybrid actors located outside of traditional forms of textuality. In some cases, the combination between distinct actors may be limited to a function of an instrumental type, but in others it can radically modify an initial action program, contributing to the formation of a new actor who, as seen earlier, is not trivially reducible to the juxtaposition of two pre-existing elements.

The paradigmatic case proposed by Latour is that of an ordinary citizen in possession of a weapon, a “gunman” whose capacity for action is not explainable in terms of the simple juxtaposition of a human actor and of a technologically advanced object, but rather in terms of the formation of a new type of hybrid actor, the result of a symmetrical translation process (citizen-gun, gun-citizen ) that modifies the status of both heterogeneous entities upstream of the association:

A good citizen becomes a criminal, a bad guy becomes a worse guy; a silent gun becomes a fired gun, a new gun becomes a used gun, a sporting gun becomes a weapon. The twin mistake of the materialists and the sociologists is to start with essences, those of subjects or those of objects. That starting point renders impossible our measurement of the mediating role of techniques. Neither subject nor object (nor their goals) is fixed (Latour, 1994: 33).

In this sense, unlike the theory of narrativity in which programs are clearly attributable to specific actors who occupy a precise position in the articulation of the narrative (syntax), for ANT action programs cannot be attributed to a single actor or narrative role but necessarily refer to an actor-network, an association between different agents variously connected to each other, a collective that from an ontological point of view represents more than the sum of its individual components.

In this network conception of agency, a decisive role is also assigned to the ability of an actor, for example an object, to act as a mediator, contributing to the formation of hybrids. To explain this process, Latour (1991b) uses the description of the functioning of a singular object: the Berlin key. The story is that of a condominium doorman whose main program could be summarized in the form of an utterance like: “please kindly always lock the door at night and leave it open during the day”. To convince the tenants and carry out his program, the “hero” of the story can resort to a usage program, for example by pointing out to everyone the annoyance of having to constantly check if the door is locked, or alternatively resort to signs (attached program). Unfortunately, the success of this goal (seemingly elementary) is hindered by a large number of anti-programs that refer to a multitude of actors: thieves, doctors, postmen. To overcome the difficulty of overcoming these tests, a devised solution consists in forming an unprecedented association between the doorman, a new key of seemingly “surrealist” shape, and a special type of lock (Figs. 3.3 and 3.4).

Fig. 3.3

Key operation—street side

Fig. 3.4

Key operation—courtyard side

The Berlin key is an ingenious double model with two symmetrical ends, once inserted into the lock it cannot be removed as usual, this action is in fact forbidden, prescribed by a lock equipped with a special mechanism. Apparently, the only possibility for the subject to recover the key is therefore to pass it through the other side of the door. Even this action, however, is not enough, if the tenant gives up recovering it, he indeed loses his own competence and with it the ability to enter or exit the building. It is at this point that the subject is forced to perform the only resolving action which consists in turning the key one more time, closing the door behind him/her to finally regain possession of his/her precious “sesame”. During the day this elaborate mechanism that replaces the role of the doorman is disabled and the human subject again assumes the canonical function of controlling the flows in and out of the condominium.

The success of the initiative is due to the fact that the new key manifests the presence of a script that allows the action program to materialize ensuring that disobedient tenants close the door at night after entering the building. The story of this singular tool is useful to Latour to highlight two aspects that make the functioning of objects peculiar compared to the logic that ensures the functioning of textuality.

The first consists in the fact that an action program originally characterized by a verbal form transfers itself into a material object, assuming a materiality, thus translating into the key.

The second concerns the fact that in this translation operation mediated by the object, the agency of the key does not simply consist in expressing the same verbal content in a new form. If it were so, it would simply function as an intermediary, limiting itself to

[…] carry, transport, shift, incarnate, express, reify, objectify, reflect, the meaning of the phrase: ‘Lock the door behind you during the night, and never during the day’, or, more politically: ‘Let us settle the class struggle between owners and tenants, rich people and thieves, right-wing Berliners and left-wing Berliners (Latour, 1991b: 18).

Rather, the association of key and lock implies a transformation on the level of agency, in other words it is configured as a real mediator,⁹ a social actor who concretely realizes the preconditions that guarantee the successful outcome of the action program. The functioning of the device, in fact, is always to be understood in a relational, network perspective. The key exists (and possibly resists over time to the trials of strength to which it is subjected) only in function of the programs and anti-programs of all the other actors (human and non-human) involved. Only the association of individual actors ensures the realization of the action program, at the same time defining the meaning of the object. The failure of one of the actors involved not only implies the failure of the entire project but also the loss of the object's ability to signify. This is what could happen if someone managed to tamper with the door or the doorman forgot to disable the lock during daylight hours.

This clarification is important because it highlights that for ANT a technical object cannot be studied beyond what can be done with a human subject. Rather, the researcher always deals with a device, that is, with a concatenation of heterogeneous actants in which competences and performances are distributed.

In this perspective, explicitly in line with the anti-essentialist tradition, the actors, whether objects, concepts or processes, are therefore understood only as nodes of a network, acquire a semiotic dimension, and manifest a meaning, exclusively as a result of the relations that are established between them and that guarantee the realization of a common action program.¹⁰

Hybrid, as in the case of the gunman, is therefore one of the terms most frequently associated with Latour's work which, by virtue of its connection with the notion of actant, has not failed to arouse interest in contemporary semiotic research,¹¹ particularly regarding the strength of this concept in relation to the overall theoretical framework developed to account for the logics of production and circulation of meaning. Thus, if on the one hand the example of the gun has been invoked to recognize the originality of Latour’s reinterpretation of the role that artifacts play in establishing and regulating a whole set of intersubjective relationships (objects not only do, but make do), on the other hand it has prompted a further exploration of the peculiarities that mark the semiotic perspective in the study of the modes of associations between humans and non-humans. The word hybrid, in fact, does not belong to the metalanguage of semiotics, which rather traces the analysis of associations between multiform elements to the distinction between actants and actors, that is, as previously clarified, to the difference between a deep level of the production of meaning, marked by the presence of abstract narrative roles, and a more superficial level, in which these instances take on a distinctive configuration, become visible, assuming a human, non-human or multiform shape. Thus, where the distinction between actant and actor in Latour can sometimes appear blurred, for semiotics a hybrid must be traced back to the principle of narrativity, as evidenced by another paradigmatic case of coexistence between humans and non-humans, that of the man-cellphone, where the tool is not reduced to a technological prosthesis that allows the exercise of some communicative practoces previosly impossible, but a real actor playing social roles only partially inscribed in its initial design (my trans.; Marrone, 2002: 29). In other words, the association between humans and non-humans—in this case far from temporary if we think about the pervasiveness of smartphones in the daily life of every one of us—takes shape and consolidates over time because it necessarily refers to a deeper level of meaning, fulfils thematic and narrative roles, and gives shape to a new agent within a specific narrative situation. Just think that, if on the one hand, we consumers inscribe in the technological object a whole series of values, for example delegating to the device the function of guarding photos and messages invested with a sentimental value, on the other hand it is the artifact that acts towards us as a subject, constitutes the context of its own use, invests itself (for us and together with us) with a meaning. Perhaps, preparing by delegation a gallery of images that reminds us of special moments, bringing to our attention an object of value that solicits a cognitive and emotional reaction, inviting us to action, to share it with others, to celebrate it.

Thus, as we will see in the final part of the book, the moment we acquire a smart object today, we contribute to the construction of a new actor with an action program that brings with it a new intersubjective situation in which the humans involved will be led to experience a series of entirely peculiar relationships. This contribution is made through translation, that is, through a series of shifts and delegations involving multiple entities. Thus, if on one hand semiotics reaffirms, despite Latour’s criticisms, the full relevance of narrativity as a principle around which to account for the meaning acquired by associations between humans and non-humans (the resulting actors are liable to act as senders, subjects, helpers … but also to assume a number of thematic roles, for example a smart watch promises to be our best personal trainer), on the other hand the relevance of his reflection on hybrids is anything but irreconcilable with the overcoming of the distinction between text and context employed by sociosemiotics itself. As Marrone (my trans.; 2002: 30–31) clarifies: “the textual nature of the object, its expressive and semantic configurations are not inscribed in the object taken in itself, as ontological properties that it possesses, so to speak, from birth, by nature or by design intent”. Rather, as we have seen and as will further develop in the following pages, the meaning of an artifact resides in the set of intersubjective and inter-objective relations that take shape from it and around it, assuming value for us. In short, Latour’s criticism of a semiotics considered excessively anchored to the canonical forms of narrativity does not take into account a very relevant turn in the discipline that has long since rethought the notion of text, understood now not as an object whose boundaries are conventionally defined, for example a novel, but rather as a model of analysis of social phenomena, in which closure is not ontologically given but constitutes one of the criteria for textuality to occur.¹² Once again, therefore, what emerges from the comparison of these two interpretive tracks is not so much an unbridgeable gap that affects the epistemological plane, but rather a methodological misalignment in the trajectories to follow to explore meaning in action. A gap that is certainly worth exploring, as we will try to do in the subsequent chapters.

3.5 From Nature/Culture to the Collective

The role played by the notion of hybrid in overcoming the distinction between subject and object is central in the work of Latour, who in We Have Never Been Modern, as mentioned earlier, aims to demonstrate the existence of a glaring paradox: that of a modernity that insists on separating nature and culture, denying the existence of increasingly numerous associations of multiform entities that itself incessantly produces.

The theoretical reflection on the crisis of modernity arises from the hypothesis that the expression “modern” refers to two types of processes that need to be clearly separated in order to function, although in reality they are strongly linked to each other.

The first type of practice manifests itself as a work of translation or of mediation that enables the association of new entities that take shape from the hybridization of nature and culture.

The second action, specular to the first, is defined, as mentioned, purification, and consists in concealing the assemblage process necessary to shape a hybrid. This work of cancellation produces as a result the formation of two distinct and apparently ontologically irreconcilable areas, one inhabited by humans and the other by non-humans.

Upon a thorough analysis, the two practices reveal themselves to be deeply connected to each other. The incessant proliferation of associations between hybrids (humans and non-humans) indeed provides the material for the action of purification which in turn, insisting on reducing the multiplicity of the elements involved within clearly distinct categories, allows the practice of translation to continue. Latour clarifies

The first set corresponds to what I have called networks; the second to what I shall call the modern critical stance. The first, for example, would link in one continuous chain the chemistry of the upper atmosphere, scientific and industrial strategies, the preoccupations of heads of state, the anxieties of ecologists; the second would establish a partition between a natural world that has always been there, a society with predictable and stable interests and stakes, and a discourse that is independent of both reference and society (Latour, 1993b: 11).

The essence and the crisis of modernity are therefore traced back to the presence of an apparently insurmountable separation between two dimensions: that of science and that of politics, to a forced dissociation in which the representation of nature, which manifests itself characteristically in the practice of the laboratory, appears destined to be inexorably separated from the political representation guaranteed by the social contract. The paradox of modernity is expressed, consequently, also in the “crisis of critical stance”, in the difficulty of offering a perspective of exhaustive analysis of contemporaneity, in other words in an impasse that manifests itself once again in the unbridgeable distance between three different interpretative approaches to the world: nature, politics, language.

The first (naturalization), typical of the so-called exact sciences, aims to account for the so-called naturalized facts, in this way excluding the social dimension of phenomena and the role played by the languages used to provide a representation; with the second (socialization) the perspective instead shifts to the critical study of the social dimension of power, relegating to the margins, or completely erasing, science, technique and the dimension of content; the third repertoire of the critique of modernity, finally (deconstruction), explicitly concerns the investigation into the meaning that, by highlighting the dimension of language, representation, text, and discourse, would end up ignoring the nature of phenomena and the social context in which they manifest and circulate.

Each of these critical approaches, while undeniably expressing its own strength, appears irreconcilable with the others. It is from this observation that the project to explore the associations between facts, power and discourse moves, trying to re-tie a “Gordian knot” around the notion of network, understood (recall the example of a technical object like the Berlin key), in terms of a translation process. To describe the peculiarity of networks, Latour significantly resorts to some metaphors of space, clarifying how they allow to cross the “borders of the great fiefdoms of criticism: they are neither objective nor social, nor are they effects of discourse, even though they are real, and collective, and discursive” (Latour, 1993b: 6). And again: (Latour, 1993b: 6–7).

[…] The tiny networks we have unfolded are torn apart like the Kurds by the Iranians, the Iraqis and the Turks; once night has fallen, they slip across borders to get married, and they dream of a common homeland that would be carved out of the three countries which have divided them up.

Following the networks, undertaking a meticulous and systematic description allows to grasp their entirely peculiar characteristics that make them “simultaneously real, like nature, narrated, like discourse, and collective, like society” (ibid). Reconstructing the logics of mediation and delegation that regulate the association between multiple actors thus produces the outcome of revealing the fallacy of a great double division, which opposes nature and culture on one hand, and the so-called moderns and the premoderns on the other.

Now, Latour recalls, recovering the fundamental work of Lévi-Strauss (1952), what characterizes the premoderns is precisely the ability to develop a monistic conception of their own nature-cultures, avoiding acting, unlike the moderns, by dissociation.

The native is a logical hoarder: he knots threads together without rest, indefatigably folding all the aspects of the real, whether physical, social, or mental, upon themselves. We traffic in our ideas: he hoards them as his treasure. Wild thought puts into practice a philosophy of finitude (Lévi-Strauss, 2021: 303).

For the premoderns, the presence of a mixture between natural and social order makes it impossible to modify the former without a change in the latter. Since “every monster becomes visible and thinkable and explicitly poses serious problems for the social order, the cosmos, or divine laws” (Latour, 1993b: 42), this results in an attitude of utmost caution. Completely different is the case of the moderns, defined as “victims of their own success”, unable to account for the increasingly evident short circuit generated by the hybrid actors taking shape from the encounter of multitudes of humans with a nature no longer distant and dominable—as in the case of the ozone hole or the greenhouse effect.

Where are we to put these hybrids? Are they human? Human because they are our work. Are they natural? Natural because they are not our doing. Are they local or global? Both. As for the human masses that have been made to multiply as a result of the virtues and vices of medicine and economics, they are no easier to situate. In what world are these multitudes to be housed? Are we in the realm of biology, sociology, natural history, ethics, sociobiology? This is our own doing, yet the laws of demography and economics are infinitely beyond us. Is the demographic time bomb local or global? Both. Thus, the two constitutional guarantees of the moderns—the universal laws of things, and the inalienable rights of subjects—can no longer be recognized either on the side of Nature or on the side of the Social. The destiny of the starving multitudes and the fate of our poor planet are connected by the same Gordian knot that no Alexander will ever again manage to sever (Latour, 1993b: 50).

Admitting that we have never been modern means then to operate a counter-revolution, accepting the challenge of overturning an apparently indisputable perspective. The poles of the object and the subject/society will no longer be understood as the irreplaceable assumptions to anchor the explanations of contemporaneity, but rather as a meaning effect (stabilization). In other words, the result of a ceaseless practice of purification that masks the networks of mediation and translation in which different actors operate and which will become the real field to explore, the terrain in which to venture.Since the entities (intermediaries and mediators) involved in the construction of the collective can assume different modes of existence¹³ in the trajectory of the associations in which they are involved, to account for the mixtures that shape the hybrids it becomes essential to connect the separate perspectives of interpretation of modernity. Among these, a decisive role is once again assigned to the pole of discourse, therefore to semiotics, but on condition of considering it no longer separately, as a world unto itself, but rather as “a population of actants that mix with things as well as with societies, uphold the former and the latter alike, and hold on to them both” (Latour, 1993b: 90).

The recourse to semiotics is manifested again with clarity and, once again, is accompanied by a critique that does not concern the legitimacy of the discipline on the epistemological plane, but rather the perimeter of its radius of action and the ways of its application. In particular, as highlighted previously, Latour’s objection focuses on a conception of textuality considered reductive because excessively centred on the model of natural language, which would prevent recognizing the narrative dimension inherent in the functioning of the real world, in other words to understand things as narratives and at the same time texts as tools of social ties. However, note again that Latour advances these critical considerations in the early Nineties, that is, in a phase in which the discursive turn of semiotics, aimed at extending the notion of text from object to model of analysis and examining social phenomena understood as multifaceted discourses, had not yet consolidated. In any case, in the project of critical analysis of modernity, the decisive role assigned to ethnology is reiterated, considered an essential perspective to trace the incessant work of translation carried out by the actants. Latour envisions the contribution made by an anthropology “returning from the Tropics”, capable of applying to the modern world the principle of generalized symmetry that allows the researcher to take a position in the intermediate space that unfolds between the poles of nature and culture, a strategic point to observe the practices through which multiple actors take shape, endowed with human and non-human properties. In this perspective, for example, a Marc Augé would not consider only “[…] some graffiti on the walls of subway corridors”, but “the sociotechnological network of the metro itself: its engineers as well as its drivers, its directors and its clients, the employer State, the whole shebang—simply doing at home what he had always done elsewhere” (Latour, 1993b: 100–101).

The recourse to anthropology here assumes a strategic value as it highlights how the separation between a universal nature (mononaturalism) and a multiplicity of cultures (multiculturalism) is overcome by the observation of the functioning of networks composed of hybrids. By observing the mediation practices of the “moderns”, one can detect how the assemblages of human and non-human actors are the result of natures-cultures that are precisely called collectives. This term is introduced to reiterate that the outcome of the association between distinct entities does not coincide with the sociological notion of society that focuses on the ties between humans, nor with that of nature elaborated in the context of epistemology and that concerns things in themselves. In the perspective of symmetric anthropology, both the moderns and the non-moderns cannot refrain from summoning and redistributing elements of the natural world and entities of the social. In continuity with what was highlighted in the work on science in action, what changes is the number of components involved, the properties that are assigned to them, the associations considered acceptable, and the extension of the networks they become part of.

At the beginning of the weighing-in process, a nuclear power plant, or a hole in the ozone layer, or a map of the human genome, or a rubber-tyred metro train, or a satellite network, or a cluster of galaxies, weighs no more than a wood fire, or the sky that may fall on our heads, or a genealogy, or a cart, or spirits visible in the heavens, or a cosmogony (Latour, 1993b: 108).

The differences, in other words, concern the degree of mobilization of the elements involved, the size of the collectives that in “modernity” are characterized by a proliferation of non-humans and the increasingly close relationship they maintain with humans. From this perspective, what distinguishes the moderns is therefore the invention of particularly extensive networks, the considerable breadth of associations between distinct entities, recruited in places very distant from each other, but always endowed with the ability to produce a transformation in the translation process in which they are involved, acting as actants.¹⁴ The profound rethinking of the notion of agency on which Latour’s thought is based thus translates into the proposal of a new model of democracy conceived to respond to the need to offer representation to the collectives that shape contemporaneity, a new constitution tasked with “replacing the mad proliferation of hybrids with their regulated and jointly decided production”. Essentially a democracy extended to the missing masses of sociological thought: the artifacts.

Recognizing the existence and proliferation of a multitude of non-human agents is therefore an essential step to explore “the enigma of their association”, to attempt to account for the way in which the very idea of nature inevitably implies a political value. Latour does not fail to reiterate this¹⁵ in Politics of Nature, a famous work aimed at claiming once again the urgency of recognizing the existence of the concatenations of humans and non-humans, with the aim of guaranteeing a representation understood in the proper sense of political sciences, that is, as a space for deliberation capable of sanctioning their reunification in view of the desired participation in a future common world.

As soon as we add to dinosaurs their paleontologists, to particles their accelerators, to ecosystems their monitoring instruments, to energy systems their standards and the hypothesis on the basis of which calculations are made, to the ozone holes their meteorologists and their chemists, we have already ceased entirely to speak of nature; instead, we are speaking of what is produced, constructed, decided, defined, in a learned City whose ecology is almost as complex as that of the world it is coming to know (Latour, 2004: 35).

The collective is therefore conceived as the new political model tasked with accounting for the cohabitation of human and non-human members, as an organization tasked with verifying the possibility of composing a world capable of associating multiform agents. Significantly, once again, this term is distinguished from that of society, considered inadequate because it is based on the stubborn distinction between the world of objects and the world of subjects, a separation that would hinder the description and understanding of the dynamic procedures of association between composite actants.

In this perspective, therefore, the collective does not designate an already constituted reality, it should not therefore be understood in the singular, rather it is a procedure to test the multiple associations between candidates for action.

The double operation that consists, on the one hand, in the recourse to a theory of relational and non-anthropomorphic action and, on the other, in its extension from the realm of the narrative to that of the social sphere, thus highlights how non-humans can be considered in all respects as social actors. Provided that we abandon the strict laws of causality on which the prejudice that reduces non-humans to simple objects is based and recognize, at least initially, the status of an uncertain, provocative entity.

Actors are defined above all as obstacles, scandals, as what suspends mastery, as what gets in the way of domination, as what interrupts the closure and the composition of the collective. To put it crudely, human and nonhuman actors appear first of all as trouble- makers. The notion of recalcitrance offers the most appropriate approach to defining their action (Latour, 2004: 81).

What guarantees the formation and maintenance of a hybrid actor is therefore the transfer of agency between different elements, the set of delegations and mediations that allows a relationship to be established between heterogeneous entities endowed with different modes of existence. In Latour’s thought, as we will see in the next chapter, this process is defined as enunciation.

3.6 The Social as Association. From Utterance to Propositions. The Principle of Articulation

Further proof of Latour’s aptitude for employing and extending notions elaborated to account for the functioning of languages is the term proposition, a keyword used to account for the peculiar modalities that characterize the associations between multiform entities within a collective. In Politics of nature the ways that allow to start a “civic work of collection”, that is, a set of practices capable of potentially ensuring the reunification of multiple actors, are examined.

The choice of this term is justified by the need not to resort to the notion of utterance, considered insidious because it is thought, once again, to be reducible to the separation between language and world, to the irreconcilable detachment between the domain of discourse and that of facts which, as seen in the previous sections, constitutes an illusion rooted in modernity. Behind this choice there is thus, once again, a critique of the mononaturalism-multiculturalism dichotomy, of the idea of an abyssal detachment between a single world and a multiplicity of languages, united by a bridge as thin, as ineffective, that of reference.

In the perspective of the political ecology explored by Latour the focus shifts rather on the need to reunify distinct entities characterized by the capacity to produce a transformation, giving shape to a network-type agency that escapes a rigid articulation, constituting itself rather as distributed both in space and in time. The term proposition is therefore no longer understood in the canonical terms of the philosophy of language, that is, as the designation of a being of the world or of a linguistic form, but rather in terms of a proposal that allows an unprecedented association of humans and non-humans to be taken into consideration in the process of forming a collective.

In this sense, the proposition, deprived of its anthropomorphic connotation, represents the way in which the instances of a civil cohabitation characterized by “variable degrees of reality” (2004: 83) manifest themselves, as in the case of “a river, a troop of elephants, a climate, El Niño, a mayor, a town, a park”, which express the presence of needs which are different but strongly linked to each other (ibid).

According to this perspective, which clearly extends well beyond the dimension of natural language, the proposition allows both human and non-human actors to manifest their presence, to bring out new modes of association, unpredictable concatenations of hybrids composed of an implementable repertoire of different elements, potentially open to new entries, capable of enriching and at the same time complicating the overall articulation of the whole. The proposition is therefore considered as the opposite of an utterance, here understood in terms of an assertive affirmation, as it consists in the proposal of a mode of existence that allows an entity to manifest its own contribution within a phenomenon of wide scope, the outcome of which is not foreseeable a priori.

If an actor is therefore defined as anything that modifies another within a trial, expressing its own agency in the capacity to associate with other entities, its proposition coincides with the elementary actions it is able to exercise in the set of concatenations in which it is involved. While utterances are judged on the basis of their reliability, that is, the truth or falsehood they express, the criterion for accounting for propositions is the soundness of their concatenation, which is defined as articulation.

Once again, the reference to linguistics is clear, however Latour reiterates how the choice of this term is also dictated by its current use in other fields such as anatomy and law, covering a wide range of connotations that are also valuable in accounting for the “insistent reality of material things” in determining social phenomena.

We shall say of a collective that it is more or less articulated, in every sense of the word: that it “speaks” more, that it is subtler and more astute, that it includes more articles, discrete units, or concerned parties, that it mixes them together with greater degrees of freedom, that it deploys longer lists of actions (Latour, 2004: 86).

Since it is no longer possible to recognize the distinction between a nature and several cultures, the good articulation of a collective passes through the ability to recruit a collection of potential actors whose integration must in any case be put to the test, in the set of propositions that they are able to manifest and that testify, only a posteriori, their competence. Overcoming the seemingly insurmountable opposition between the two poles of subject and object, the term social thus gives way to that of association. The utterance understood as a component of human language that seeks to verify its own adequacy to the world of objects through an operation of reference, is therefore reimagined as what is produced by the collective interrupted in its exploration of the common world. This change of perspective thus shifts the focus onto the circulation of agency, that is, onto the ability to effect a transformation by passing through different modes of existence, in other words, on the enunciation.