Agents and kinding

How does a particular sign come to stand for an object instead of some other sign and why does it give rise to a particular interpretant and not some other interpretant? What directs the unfolding of semiotic processes? To supplement the general model of meaning introduced above and to bring it closer to the object of this study, let us now discuss agency and introduce the notions of selection and sieving (see Kockelman 2013a: 17–19, 41–42, 81–85;

on the relation of Peircean semiotics to agency, see also e.g. Ransdell 1986:

54; Colapietro 1989: xix, 95–97).

An agent is whatever is capable of affecting which interpretant a sign gives rise to. It may be a selecting agent that is capable of sensing a sign and instigating an interpretant. In the case of humans, the agent may be more encompassing than an individual person (e.g., a group) or less encompassing (e.g., a mental state). Kockelman (2013a: 20–21) gives an example of a stereotypic enchaining of cognitive processes: an object first causes a sensation; the sensation, then, indexically (or causally) gives rise to a perception, the perception inferentially (or logically) to a belief, the belief to an intention, the intention to an instigation of, say, a speech act (see also Rosenthal 2004). Each state in the chain can be framed as an agent, which can itself be entangled in many other semiotic processes. That is, agents are semiotic processes too and, therefore, fundamentally distributed by nature (see 2.1.5) and only under particular framings coincide with entities such as persons.

The agent may also be a sieving agent that “gives rise to consequences for no other reason than serendipity” (Kockelman 2013a: 29, original italics).

For instance, the features of a particular environment, instrument, or infrastructure may limit the the unfolding of semiotic processes in particular

ways. Any semiotic process involves both selection and sieving. In fact, a major source of sieving are those processes that are already in place or under way (e.g., previous interpretations or presumed cultural norms, learned codes, habitual channels or infrastructures) (see Kockelman 2013a: 44).

Agents operate on different time scales. Any single framing tends to overly reify agency. What, for instance, on an interactional time scale appears to be a relatively intentional selection by a person, may appear as non-intentionally sieved or selected for on evolutionary, sociohistorical, or biographical time scales (e.g., a genetic predisposition caused by natural selection or a relatively arbitrary habitual behavior one has been socialized into).

Let us now have a look at the following diagram (2). The top half of the tetragon (S-O-I) corresponds to the sign-object-interpretant relations familiar from diagram 1. The bottom half represents the sign-agent-interpretant relations. The agent (A) takes the sign as “input” and yields as the “output” an interpretant. In other words, the agent interprets the sign-object relation according to its interests or characteristics.

Diagram 2 Relations between relations based on Kockelman (2013a).

As was mentioned earlier, any component in a semiotic process can simultaneously be a component in other semiotic processes. Choosing among different possibilities (i.e., seeing a particular component as a sign, object, or interpretant in a particular process) is referred to as semiotic framing (see Kockelman 2005: 236, 269–271; 2013a: 50). For instance, in the previous diagram, the interpretant (I1) of the first process is a new sign (S2) in the subsequent process.

We can illustrate framing in light of the object of this study. Any segment of an entextualized online dating advertisement can be framed both as a sign and as an interpretant. When a writer describes some aspect of himself – think of it as O1 in the diagram – such as a behavioral routine, a physical attribute, or an emotional pattern, which he knows through some sign(s) (S1) (e.g., feelings, memory images, perceived reflections in the mirror), the act of giving such non-linguistic signs an explicit linguistic description functions as

… I2

A2

A1

S1

O2

O1

I1=S2 …

a representational interpretant (I1). It formulates the object as a particular kind of semiotic object by projecting propositional and discursive structure onto it. That is, the interpretant interprets the sign in a particular way in light of particular cultural symbols (and with particular consequences in terms of possible inferences, accountability, etc.). At the same time such linguistic descriptions function as addressed signs (S2) to be interpreted by (anticipated) readers (A2). They eventually prompt new interpretants (I2) from those interactants that actually end up reading the advertisement (e.g., mental representations of the writer as a person; feelings of suspicion, indifference, trust, interest, desire; the decision to reply or not; the reply text).¹⁷ Self-presentation in dating advertisements, then, is a dialogue both between writers and readers and between past and future versions of the writer. As an interpretant, the text interprets both for self and for others those signs that serve as ingredients for the text.

From the standpoint of agency, it is noteworthy that the writer, as agent A1 that instigates the representational interpretant (I1), will be able to anticipate the readers’ (A2) interpretations (I2) and can try to signify in a particular way in order to bring about a particular kind of interpretation.

Addressed signs are the kinds of signs that have been expressed for the sake of particular interpretants to which the signer is committed (Kockelman 2005: 252; see also e.g. section 5.4). That is, writers interpret themselves (I1) in ways that are addressed to actual or imagined others and are meant to prompt desirable kinds of interpretations (I2). In terms of interactional turn-taking, one may speak of the mobilizing of particular kinds of responses from others (see e.g. Stivers & Rossano 2010). The capacity of agents (1) to control the expression of signs, (2) to compose particular kinds of object-sign relations, and (3) to commit to particular kinds of interpretants is called practical agency. Their capacity to representationally (1) thematize and (2) characterize semiotic processes and (3) to reason about them is called theoretical agency. (See Kockelman 2013a: 81–82.) Agents, then, have various degrees of flexibility (practical and theoretical agency), for which they are made accountable by others’ recognizing and regimenting interpretants (e.g., in the form of evaluation, entitlement, and obligation, see Enfield 2013). (Specifically from the standpoint of this study, see also chapter 3.)

17 The interpretants produced by the readers are additional interpretants in the first process, since they relate to the same object (i.e., I2 mediately interprets O1). One might say that they point to the same dynamic object that becomes represented as two different immediate objects for two different persons. To what extent the writer’s knowledge of the object (O1) and the reader’s knowledge of the same object (O2) come to correspond to one another is a matter of the intersubjective calibration of the semiotic processes. Such questions are, of course, at the heart of all social interaction but, as will be seen later, they can find particularly acute forms in settings like online dating advertisements in which the interactants are physically displaced from one another and may sometimes hold very different interests.

To return to the question of selection and sieving in online dating advertisements, it should be noted that the conscious choices made by writers of online dating advertisements in terms of what to reveal and how is merely one of the most apparent manifestations of agency. A wide range of semiotic processes can occupy the position of a sieving or selecting agent. For instance, various cultural norms and ontologies, including genre models and models of personhood, both enable and constrain particular forms of signifying and interpreting. Similarly, habitual and less conscious patterns of self-presentation that individual writers have developed shape the composition of their texts. Furthermore, the characteristics of the channel and the instrument that serve as infrastructure for the linguistic self-presentation considerably sieve the range of possible object-sign and sign-interpretant relations. As will be seen in more detail later, the fact that a digital text-artifact can only carry certain kinds of signs that can only be interpreted in certain ways will be of utmost importance to the social interaction they mediate.

Finally, let us introduce a few more essential concepts that will be central later on. A reader who stumbles across an online dating advertisement on a dating forum text becomes, first of all, aware of the fact that some actual human individual exists out there, based on some very basic indices (such as the fact that someone has written and posted the text). Moreover, the reader can instantly project upon that individual some basic cultural default understandings of what persons generally speaking are like. Then, by interpreting whatever signs are perceivable for the reader in the text-artifact, he or she can project more specific kinds, such as social statuses, mental states, or physical characteristics, on that individual, gradually building a more specific interpretation of the person as a semiotic object. From a more general perspective, individuals, however, need not be human individuals and kinds need not be human kinds. A kind is any projected propensity to exhibit particular signs, or to “behave” in particular ways (and is, therefore, one type of ultimate interpretant). An individual is any relatively stable background (or, in some sense, a relatively enclosed object) on which more specific (and sometimes less stable and less enclosed) kinds can be projected as semiosis unfolds. As will be seen in 2.2, the text(-artifact) itself is an individual on which various kinds (such as the ascription of a particular genre) can be projected based on various indices. Those assumptions about indices, kinds, and individuals that enable the interpreting agent to perceive indices and to project kinds on individuals are called a semiotic ontology.

(See Kockelman 2013a: 4–6, 54–56; 2013b.) The same signs can be read as indices of different kinds in light of different ontologies. For instance, the characteristics that are projected on writers based on the signs they exhibit in their texts depend considerably on the readers’ ontologies and can result in widely different interpretations, as will be seen in 4.2.