Artefacts as Signs, Signs as Artefacts

2.1 Language/linguistic production

First of all, let us consider the pair language/linguistic production. In his 1977 monograph, Linguistics and economics, Rossi-Landi specifies that the term language – and, consequently, the adjective “linguistic” – refer to the verbal sign systems, more specifically, to “verbal (oral and written) codes and messages” (Rossi-Landi 1977: 17). Of course, he is well aware of the fact that this kind of connotation could be “rather conservative” (Rossi-Landi 1977: 17), but he also underlines that such a terminological choice depends on his will to distance himself from semiotic theories which assign priority to the verbal sign system precisely, that is, from those theories which assert that nonverbal sign systems depend on the verbal. Hence, distinguishing language from nonverbal sign systems, Rossi-Landi intends to recognize

[their] reciprocal structural independence and at the same time [their] reciprocal influence. This does not mean that there was no hierarchy imposed by real needs on sign systems;but it does mean that all basic social sign systems are ‘primary modeling systems’ – according to the terminology of Lotman, who considers primary only the system of language, and secondary instead all non-verbal sign systems. (Rossi-Landi 1977: 18)

Let us focus on this latter concept. Rossi-Landi is referring to the theory of Jurij Lotman (1922–1993) – the founder of Moscow-Tartu School of Semiotics – and he is criticizing the fact that this theory only considers verbal language as a primary modeling system. To explain this argument, it is necessary to briefly refer to the theory of Thomas Sebeok (1920–2001).

Modifying the notions of primary and secondary modeling systems as structured by the Moscow-Tartu School, Sebeok develops a modeling systems theory (Sebeok and Danesi 2000) based on three levels: primary, secondary, and tertiary. According to Sebeok, the primary modeling system “allows organisms to simulate something in species-specific ways” (Sebeok and Danesi 2000: 45) and the human animal shares such a device with the other species; indeed, every living being uses its “sensorium” (Sebeok 2003: 174) – that is, the totality of its senses – to structure a sign system through which an interpretation of – and a life in – the external word can be possible.

Nevertheless, the human species distinguishes itself from others exactly in its typical and innate capacity for simulative modeling. This typical and innate capacity is language, whose fundamental property consists in what Sebeok defines as syntax (Sebeok 2003: 178), that is, the ability to generate an indefinite number of meaningful signs, using a limited number of “construction pieces” (Ponzio and Petrilli 2004: 215). This means that human beings by means of their primary modeling system can attribute to the same object a potentially infinite number of meanings.

From such a perspective, the term “language” does not refer to the verbal sign system, but rather to “a mute verbal modeling system” (Sebeok 2003: 175) which Homo habilis – the first species of the genus “Homo” – “must have had lodged in its brain, but it could not encode it in articulate, linear speech” (Sebeok 2003: 175). Indeed, it will only be with the appearance of Homo sapiens that this modeling system is externalized and speech arises as the secondary modeling system.

According to Sebeok, “language evolved as an adaptation, whereas speech developed out of language as a derivative ‘exaptation’” (Sebeok 2003: 176); this means that, in the course of the hominization process, language “was built by selection for the cognitive function of modeling” (Sebeok 2003: 176), and only with the appearance of the Homo sapiens did such a primary system “[come]to be ‘exapted’ [that is, transformed in its function] for communication, first in the form of speech (and later of script, and so forth)” (Sebeok 2003: 176).

Thus – according to Sebeok – becoming a communication-oriented device, language enhances the nonverbal capacity with which human beings were already endowed, allowing them to operate both on the verbal and nonverbal level (see Sebeok 1998: 25); in this way, using this enhanced capacity, human animals can structure their tertiary modeling systems; and, from these highly abstract, symbol-based modeling processes, human cultural systems finally emerge.

From such a perspective, Sebeok can affirm that what the Moscow-Tartu School calls primary, that is, the verbal sign system, is “phylogenetically as well as ontogenetically secondary to the nonverbal; and, therefore, what they call ‘secondary’ is actually a further, tertiary augmentation of the former” (Sebeok 2003: 175).

In the light of all these theses, it should be clear that, when Rossi-Landi asserts that verbal and nonverbal sign systems are both primary modeling systems, he draws attention to the fact that the syntactical capacity is incorporated – as Sebeok (2003: 178) would say – in both of them. To explain such a thesis, Rossi-Landi takes as an example a common human utensil: a table knife.

According to Rossi-Landi a knife – like every utensil – has “the level of complexity and the utility […] of a proper sentence […]. The utensil in fact consists in parts which are produced and put together according to certain rules, like the sentence” (Rossi-Landi 1977: 22). Furthermore, every use of a certain utensil – or, as I intend to illustrate, every work that could be done with that utensil – is equivalent to everything with which the utensil “can be put into a real relation” (Rossi-Landi 1977: 22). This means that a kind of “language of things” (Rossi-Landi 1977: 22) actually exists, and this language coincides with a certain system of syntactic relations between utensils and other things; e.g. relations between utensils and materials, or between utensils and other utensils. These set of relations can be considered as a “nonverbal sign system” (Rossi-Landi 1977: 22).

This latter assumption introduces the other pair of concepts implied in Rossi-Landi’s homology: work and material production.

2.2 Work/material production

Emphasizing that every utensil is syntactically structured – just like a sentence – and that different utensils could be syntactically related according to certain uses – just like verbal discourses – Rossi-Landi maintains that both verbal and nonverbal sign systems could be analyzed in the light of the Marxian conception of work: more precisely, the human capacity to generate, communicate, and interpret articulated sounds can be understood as linguisticwork, ^[3] a kind of work representing a further genetic development of human capacity to generate, communicate, and interpret meaningful signs in general (hence, nonverbal signs also); the Italian semiotician defines this capacity as “sign work.”

In this regard, Rossi-Landi suggests that André Leroi-Gouhran’s (1911–1986) research could confirm the genetic precedence of the sign work in the hominization process. Indeed, the French paleontologist maintained that, from the most primitive technical levels, construction of utensils is planned; namely, according to Leroi-Gouhran, all the operations through which an artefact is constructed are executed with a view to its future utilization (Rossi-Landi 2006 [1985]: 228).

In semiotic terminology, we could say that the process through which human beings design and produce artefacts coincides with the abstractive process through which human beings interpret their needs; therefore, production of artefacts should be understood as a fundamental manifestation of our species-specific interpretative work. This means that every human artefact should be interpreted as a medium – an interpretative medium – to satisfy a certain human need. As Marx would say: whether the need arises “from the stomach or from imagination, makes no difference” (Marx 2002: 3).

Nevertheless, before proceeding with the analysis of the concepts of work and material production, another terminological clarification is necessary. As we know, in a footnote to the fourth German edition of Capital (Vol. 1), Friedrich Engels (1820–1895) proposes a terminological distinction between work and labor: “the English language has the advantage of possessing two separate words for these two different aspects of labor [Arbeit]. Labor which creates use-values and is qualitatively determined is called ‘work’ as opposed to ‘labor’; labor which creates value and is only measured quantitatively is called ‘labor’, as opposed to ‘work’” (Engels in Marx 2002: 2254, note 16).

In line with this distinction, Rossi-Landi declares his preference for work as a “fundamental term” (Rossi-Landi 1977: 6, note 1) for two reasons: a) because “generic work, productive of values, [should be] considered as a derivation from” (Rossi-Landi 1977: 6, note 1) the work producing use-values; furthermore, the idea of linguistic work defines “specific” (Rossi-Landi 1977: 6, note 1) work producing a peculiar type of artefacts, i.e. verbal messages. The other reason b) is strictly connected with a translation problem: “‘work’ can be used together with ‘worked’ and ‘works’, while the same derivations from labor would bear different meanings” (Rossi-Landi 1977: 6, note 1).

Nevertheless, Rossi-Landi prefers the term work even to designate concepts which refer to a “more generic activity” (Rossi-Landi 1977: 6, note 1). Indeed, explaining the possibility of applying labor value-theory^[4] to the linguistic sphere, Rossi-Landi confirms this terminological choice, defining “abstract human labor” (Marx 2002: 44, emphasis mine) as “undifferentiated work” (Rossi-Landi 1977: 52, note 12). The same can be said of the term Arbeitsprozeß. This concept is usually translated as “labor process” (see, e.g., Marx 2002: 448); but instead, Rossi-Landi renders “Arbeitsprozeß” as “working process” (Rossi-Landi 1977: 39). ^[5]

The term labor process designates the “process by which man mediates, regulates and controls his metabolism with nature through his own actions” (Marx 2002: 450). More specifically, this concept refers to the purposeful activity [zweckmässige Tätigkeit] by which human beings produce use-values – i.e. instruments and goods – to satisfy their needs, transforming the environment in which they live. Obviously, this species-specific human ability can be observed in every form of society and in every historical epoch. Thus, it is possible to affirm that the labor process is a factor which all societies have in common. Indeed, as Marx states,

The labor-process [Arbeitsprozeß] […] is purposeful activity [zweckmässige Tätigkeit] with a view to the production of use-values, appropriation of natural substances to human requirements; it is the necessary condition for effecting exchange of matter between man and Nature; it is the everlasting Nature-imposed condition of human existence, and therefore is independent of every social phase of that existence, or is common to every such phase. (Marx 2002: 472)

2.3 Language and work: A dialectical relation

All these terminological and categorical clarifications contribute to explaining how Rossi-Landi posits the dialectical (and homological) relation between language and work.

According to Rossi-Landi, language is work because human (verbal and nonverbal) signs do not exist in nature (see Rossi-Landi 1977: 31–32); therefore, they should necessarily be products of human mediating and transformative praxis, i.e. products of human labor. Furthermore – as Rossi-Landi underlines – the practical-communicative use of language presents the same elements composing the “labor process” as defined by Marx (2002: 453): the activity conforming to a goal [zweckmässigeTätigkeit], the object [Gegenstand] – i.e. the material – on which work is applied, the work instrument [Mittel], and the product [Produkt] of work. From such a perspective, language can be understood as a) work oriented by signs, i.e. work conforming to certain rules (e.g. grammatical, syntactical, pragmatic, and so on) and oriented toward specific goals (e.g., communicative goals); b) work performed on signs, constituting the material on which work is applied; and c) work performed by means of other signs, constituting the work instruments. To sum up, it is the work that every human being performs when s/he produces – or uses – a certain sign to interpret – or, more specifically, to talk about – another sign.

On the other hand, the fact that work is goal oriented, i.e. purposeful activity [zweckmässige Tätigkeit], implies the presence of language. In fact, that goal is actually a sign, that is – as Rossi-Landi maintains – “a conscious or unconscious, desired or endured, ‘mental’ anticipation of the product” (Rossi-Landi 1977: 40), namely, a design or a project. Without language – understood in its verbal and nonverbal dimension – no kind of goal-oriented activity could take place. Without meaning-making, communicative, and interpretative processes, no kind of work project or cooperation in the workplace would be possible. This is the reason why language “determines the finalistic character of work, its taking place according to a program” (Rossi-Landi 1977: 40, my italics); and a program is always no less than an organized part of a certain sign system, that is, a set of instructions which are more or less implicit in every labor process. Therefore, work is inherently linguistic, or semiotic at least. In this sense, work is language.

3.1 A Peircean approach to the artefact

We have seen that every artefact – or, adopting Marxian terminology, every use-value – is produced and used to satisfy – or, semiotically speaking, to interpret – a certain human need. Therefore, the dialectical relation between production and use (i.e. consumption) of artefacts can be understood as an interpretative process. Nevertheless, to understand concepts like interpretation, or interpretative process, we must refer to a fundamental theory, that is, Charles Sanders Peirce’s (1839–1914) semiotics. In this regard, it would seem necessary to start with the Peircean definitions of sign, underlining the fact that – according to Peirce – every sign is characterized by a triadic structure. More precisely, the sign may be understood as a continuum in which some discrete parts can be identified. These are a) the sign – or, more precisely, its representamen, b) the object, and c) the interpretant. Actually, Peirce defines a sign “as anything which is so determined by something else, called its Object, and so determines an effect upon a person, which effect I call its Interpretant, that the latter is thereby mediately determined by the former” (Peirce 1908: SS 80-81).

According to the Italian semiotician Giampaolo Proni (2002, 2012), artefacts can be analyzed in the light of this threefold partition.

What we could define as the body – or, as Proni specifies, the “material quality” (Proni 2002: 40) – of the artefact coincides with its representamen.

The object of the artefact describes how to use that material quality in order to obtain certain general goals (see Proni 2002: 40). From this perspective, an artefact can be understood as a sign characterized by a particular object: this object is a practical function. This means that an artefact is an artefact if “such a function can actually be performed” (Proni 2002: 40). An artefact, i.e. an object of use, must function, it must work.

What to understand as the interpretant of the artefact is yet to be explained. According to Proni, the interpretant is the response which the function induces in the user – that is, in the interpreter of the practical function. More precisely, this response coincides with “appropriate behavior” (Proni 2012: 41, my translation) toward the artefact. This means that the producer of the artefact – a single person or a specific organization – plans that the user should perform a specific sequence of actions in using the artefact. This sequence of actions coincides with the behavior considered as appropriate.

In this regard, interesting to underline is that Rossi-Landi explicitly connects the concept of use with the concept of behavior: “using an object means behaving toward that object” (Rossi-Landi 2006 [1985]: 161, my translation). Therefore, interpreting how the artefact should work, and how it should be used, the user interprets the specific programs inscribed in the artefact by the producer. Use of the verb to inscribe could appear as completely metaphorical, but it should be clear that the term refers to the syntactical capacity of generating – we could say, writing – meaningful signs and programming, as Rossi-Landi would say, “the behavior of pieces of matter” (Rossi-Landi 1977: 95), i.e. the capacity of organizing a specific portion of the external (living or nonliving world) according to a specific sign system.

3.2 Semiotics of assemblies

Bruno Latour and Madeleine Akrich (1992) adopt a similar model to describe what they define as assemblies – that is, the relationships between human beings and technical objects.

1. The producer – who the authors call scribe, e.g. an engineer, an inventor, a manufacturer, or a designer – in-scribes a program in a certain artefact (or technical object).

2. The artifact prescribes: that is, it allows or forbids the user certain actions.

3. The user can sub-scribe to the program of the artefact, accepting the competences and performances implied in its use. Or, conversely, the user can de-inscribe from the program, refusing those competences and performances. Moreover, the user can re-inscribe new programs in the artefact, redistributing or inventing competences and performances.

4. The analyst de-scribes the totality of this process.

In my opinion, this semiotic relation may be understood in terms of the processes of writing and reading the instructions – or, as the authors would say, the script – contained in a certain artefact.

3.3 Artefacts as systems of inferences

Probably, this dialectics between inscription and subscription/de-inscription can be further explained by adopting a reinterpretation of the Peircean theory; I am referring to the model of semiosis as structured by Charles Morris (1901–1979).

According to Morris, “the process in which something functions as sign may be called semiosis” (Morris 1938: 3). To describe such a process he constructs a set of special or technical terms, such as: sign vehicle, designatum, denotatum, interpretant, and interpreter; each of these terms expresses a relational property that a certain thing assumes “by participating in the functional process of semiosis” (Morris 1938: 4).

Semiosis is an interpretative process which takes place every time a subject – the interpreter – infers that a certain object (a sign vehicle) is related, under a certain aspect (for example, on the basis of a causality relationship), with something else (a designatum/denotatum), and in such a way that the first object is a sign of the other object. This particular relationship inheres the semantic dimension of semiosis.

The interpreter generates an interpretant when he/ she takes account of the relation between the sign vehicle and the designatum/denotatum. This further relationship inheres the pragmatic dimension. Finally, the syntactic dimension coincides with the structure of logical and formal relationships between different sign vehicles (see Morris 1938: 13–21).

These three dimensions – i.e. the three dimensions of the sign – are simultaneously involved in every process of semiosis, that is, every time a sign is generated and interpreted, the meaning of the sign – according to Morris – coincides with the system of relations between the three dimensions.

Taking account of the connection between sign vehicle and designatum, the interpreter expects that every time a certain sign vehicle appears in front of him/her, that vehicle should designate a class (see Morris 1938: 5) of certain objects, or events. If the interpreter detects one member of the class at least, the sign vehicle has its denotatum. As Morris says, “to the degree that what is expected is found as expected the sign is confirmed” (Morris 1938: 33). In this case, it can be possible to affirm that “signs are ‘true’ in so far as they correctly determine the expectations” (Morris 1938: 33) of their interpreters.

An example can clarify this latter assumption. Let us consider the utterance “there is a cat in the kitchen” as a sign vehicle. ^[6] There is a possibility that a cat could actually be in the kitchen; such a possibility constitutes the designatum of the utterance. Connecting that utterance (sign vehicle) with that possibility (designatum), the interpreter expects that a cat could be in the kitchen, and this expectation coincides with the interpretant. If the interpreter finds a cat in the kitchen, then his/her expectation will be satisfied, and there will be a denotatum for the possibility designated by the sign vehicle. From such a perspective, the utterance should be considered as true. Therefore, in line with Peirce’s semiotics, Morris supposes that every interpretative process is based on inferences (see Eco 1984: 4), and – consequently – that every sign-relation (i.e. the relation between sign vehicle, designatum/denotatum, and interpretant) is always subject to possibility and uncertainty.

I believe that this model can be appropriately applied to the semiotics of artefacts, analyzing the relation between production and use (or consumption) as a system of inferences. More specifically, on the one hand, it is possible to understand the production of an artefact as based on this inference: “if the user (interpreter) correctly follows (interpretant) the instructions (designatum), then the artefact (sing-vehicle) will properly work, and the program of use will be confirmed (denotatum).” On the other hand, it is possible to understand the use of an artefact as based on this inference: “if I (interpreter) want to obtain certain goals (denotatum) using the artefact, then I have to follow (interpretant) the instructions (designatum).”

In the light of this consideration, the object of the artefact – that is, what the artefact can possibly signify, its possible designatum – may be understood as a bidirectional process: “on one side the Object is the production process, which represents the artefact’s past, the chain of causes and effects that produced it as a material entity; on the other side the Object is the possible interaction, or use-program, that is, its future” (Proni 2002: 43).

From such a perspective, the production process and the function – that is, the use-program – of the artefact should be understood as two discrete moments of the same semiotic continuum; the body of the artefact constitutes the material event which temporarily interrupts this continuum. As Proni states: Use and production are the two directions of the causal line on which the artefact is placed as a material event. The production is the chain of causes and effects that have brought the artifact into existence. The use is the chain of possible and probable causes and effects in which the artifact will be involved. In this chain the descriptions of two subjects are inscribed: the model user and the implied producer. (Proni 2002: 48)

Proni outlines these two types of subjectivity starting from Umberto Eco’s (1932–2016) semiotics. The model user is the subject who is invoked to interpret the function in-scribed – as Akirch and Latour would also say – in the artefact by the model (or implied) producer. Adopting the categories of Eco’s semiotics, Proni intends to illustrate that an artefact can be analyzed as a particular form of textuality; indeed, just like every form of textuality, an artefact presupposes two subjectivities: a) a writer who plans and designs how the artefact must work; a subject who writes the instructions of that artefact; and b) a reader who must be able to follow the instruction written by the producer to achieve certain goals.

It should be evident that this theoretical model presents certain fundamental convergences with the semiotics of assemblies as structured by Akrich and Latour. Nevertheless, I believe that all these theses can be also analyzed in the light of Rossi-Landi’s materialistic semiotics.